Architecture of bridges

HE ARCHITECTURE OF BRIDGES

UNIVERSITY
OF FLORIDA
LIBRARIES

Digitized by the Internet Archive

in 2013

http://archive.org/details/architectureofbrOOkass

The Architecture of Bridges

THIS BOOK HAS BEEN

PRODUCED UNDER A GRANT

FROM THE AMERICAN BRIDGE COMPANY,

A SUBSIDIARY OF

THE UNITED STATES STEEL CORPORATION

Elizabeth B. Mock

T- Willis

THE ARCHITECTURE OF

Bridg

es

THE MUSEUM OF MODERN ART • NEW YORK

62V

Trustees of the Museum of Modern Art

John Hay Whitney, Chairman of the Board; Henry Allen Moe, ? st Vice-Chairman; William

A. M. Burden, 2nd Vice-Chairman; Sam A. Lewisohn, 3rd Vice-Chairman; Nelson A.
Rockefeller, President; Philip L. Goodwin, 1st Vice-President; Mrs. David M. Levy, 2nd
Vice-President; Ranald H. MacDonald, Treasurer, John E. Abbott, Alfred H. Barr, Jr.,
Mrs. Robert Woods Bliss, Stephen C. Clark, Rene d'Harnoncourt, Walt Disney, Mrs. Edsel

B. Ford, A. Conger Goodyear, Mrs. Simon Guggenheim, Wallace K. Harrison, James W.
Husted, Mrs. Albert D. Lasker, Henry R. Luce, William S. Paley, Mrs. E. B. Parkinson,
Mrs. Charles S. Payson, David Rockefeller, Beardsley Ruml, James Thrall Soby, Edward
M. M. Warburg, Monroe Wheeler.

Honorary Trustees

Frederic Clay Bartlett, Mrs. W. Murray Crane, Duncan Phillips, Paul J. Sachs, Mrs. John S.
Sheppard.

Copyright, 1949, The Museum of Modern Art, New York. Printed in the U.S.A.

CONTENTS

Page

Acknowledgments

6

Introduction

7

Structural Types

10

The Architecture of Bridges

Stone

12

Wood

30

Metal Arch

40

Suspension Cable

54

Metal Beam

64

Reinforced Concrete

84

Reinforced Concrete Arch

88

Reinforced Concrete.- Beam and Rigid Frame

112

Glossary

125

Sources of Illustrations

126

Acknowledgments

This book was started three years ago, when I was Curator of the Museum of Modern
Art's Department of Architecture, and was undertaken at the suggestion of Philip L.
Goodwin, trustee of the Museum, who for years had been urging the Department to
make an attempt to raise the level of American bridge design. What started merely
as a gesture of friendship soon became an absorbing interest, so I am grateful to Mr.
Goodwin for opening my eyes to bridges. Other sponsors were Philip C. Johnson,
now Director of the Museum's Department of Architecture and Design, and Edgar
Kaufmann, Jr., Advisor to that Department. To Mr. Johnson I am also greatly indebted
for editorial suggestions, particularly as concerned with the special problems of a
picturebook. I wish to thank Mary Barnes, former Curator of Architecture, and Ruth L.
Bookman, former Assistant Curator, for their assistance in gathering material and
their criticism of the text. The photograph reproduced on the paper jacket was sug-
gested by Edward Steichen, Director of the Photography Department.

Outside the Museum there is a long list of people who have been kind and helpful
to me in this rather presumptuous project, and to whom I wish to express abiding
gratitude: Rudolf Mock, for needed encouragement and advice as well as for his
contribution of the diagrams that illustrate Structural Types; Henry-Russell Hitchcock, Jr.,
for provocative conversation on the subject of nineteenth-century iron bridges and
the loan of relevant material; Adolf Meyer, Chief of the Tennessee Valley Authority's
Civil Design Branch, for invaluable criticism of the text; Sigfried Giedion, for the stimulus
of his long and articulate championship of the Swiss engineer, Maillart, also for the
use of some photographs which he assembled for the Museum of Modern Art's traveling
exhibition of Maillart's work; Paul Zuberbuhler, for material on old and new Swiss
bridges; Max Bill, for his photographs of Maillart bridges; G. E. Kidder Smith, for
his material on Swedish bridges; Bernard Rudofsky, for his critical reading of the
introduction; Marcel Fornerod, for information about Freyssinet's prestressed concrete
construction.

I am similarly indebted to Andre Bloc, editor of L'Arch/recfure d'aujourd'hui, Alfred
Roth, editor of the Swiss magazine, Das Werk, Waldo Bowman, editor of the Engineer-
ing News Record, Elizabeth Fitten, of Princeton University's Marquand Library, and
R. E. Enthoven, of the Library of the Royal Institute of British Architects. Other libraries
that extended special courtesies were the Technical Library of the Tennessee Valley
Authority, the Lawson McGhee Public Library of Knoxville, Tennessee, the Engineering
Societies Library of New York, the Architectural Library of Harvard University, and the
main library of Princeton University. The picture collection of the American Institute of
Steel Construction was extremely helpful in the preliminary research, and I wish par-
ticularly to thank the editors of the Engineering News Record for the loan of innumerable
photographs from their excellent files.

Above all others I am beholden to Frank Lloyd Wright, whose influence has
prompted me to search in bridges for the qualities of organic architecture. But the
interpretation is entirely my own, and I am alone responsible for its inadequacies.

Publication of the book was made feasible by a generous subsidy from the Amer-
ican Bridge Company, granted upon the recommendation of Mr. J. H. Zorn, of that
company, and Mr. Robert J. Ritchey, of the United States Steel Corporation.

E.B.M.
Taliesin West
March, 1949

ACKNOWLEDGMENTS 6

INTRODUCTION

In an old graveyard of Concord, Massachusetts, is a slab with a date, 1791, and
an inscription:

In Memory of Captain John Stone

the Architect of that Modern and

Justly Celebrated Piece of Architecture

Charles River Bridge

A similar conviction that a fine bridge is also fine— and modern— architecture, is the
basis of this book and its only justification.

Bridges are architecture, but architecture of a very special kind, unique in its single-
mindedness. Ordinarily the art of architectural or landscape design consists in the
creation of space, and structure is finally the means to that end. But since the function
of a bridge is simply the continuation of a roadway over a void, its structure is both
means and end, and its reality lies not in space enclosed, but in structure itself. Since a
bridge does not define space, but cuts through it, it is free of all the intricate psycholo-
gical considerations that must be taken into account when space is molded or enclosed.
Thus, paradoxically, a bridge is at once the most tangible and most abstract of archi-
tectural problems. As such, it is capable of extraordinary purity, though it may perhaps
never achieve the richness and depth of expression that are possible in buildings of
more complex human motivation.

Since the reality of a bridge lies in its structure, the art of bridge building lies in
the recognition and development of the beauty latent in those structural forms that
most effectively exploit the strength and special properties of a given material.

Beauty is not automatic; technical perfection alone is not enough. A great engineer
is not a slave to his formulas. He is an artist who uses his calculations as tools to create
working shapes as inevitable and harmonious in their appearance as the natural laws
behind them. He handles his material with poetic insight, revealing its inmost nature
while extracting its ultimate strength through structure appropriate to its unique powers.

Today we boast the most powerful materials of all time: steel and steel-reinforced
concrete. But there is a curious reluctance to explore their ultimate possibilities and
accept their full esthetic implications— a reluctance based on the idea that massiveness
is itself a virtue, as it was in the days when stone was the only strong, permanent,
therefore honorable material. Arch-reactionaries in this sense were the Nazis. Needing
heavily pretentious buildings to symbolize the immortal glory of their State, they de-
veloped this characteristically specious line of reasoning: "Form requires mass; without
mass, no artistic, architectural form; without form, no beauty." They cited the ruptured
Tacoma span (page 63) as evidence that "Abolishment of mass leads not only to
formlessness but to failure." German engineers in actuality paid scant attention to this
facile official theorizing, and bridge design was therefore the one field of art in which
the Third Reich was not completely tripped up by its own mock-heroics.

The identification of beauty with mass has never been as deliberate as it was in
Germany in the thirties, but the two are very often confused. Even when this confusion
is unconscious, it is a very real obstacle to the achievement and acceptance of quality
in contemporary bridge design. The old stone-builders set themselves no such limitation.
On the contrary, they were constantly seeking new ways to lighten their spans by making
every stone a lively working element of the structure, and thus to minimize the massive-
ness of what was, after all, a massive material. Man has rarely built less efficiently than
he was able, and the history of bridge architecture is essentially the story of his triumph
over space through increasingly skillful exploitation of the best materials available to
him. His triumph was not only over space, but over the inertia of material. The nineteenth-

7 INTRODUCTION

century engineers were working within this grand tradition when they welcomed the
new possibilities for efficiency— and for delight— that were offered by metal. They made
the sudden transition from the massiveness of stone to the finely etched lines of iron
and steel with magnificent assurance. Only in the last fifty years or so have bridges
been overtaken by nostalgia for the reassuring weightiness of stone construction, and
their forms falsified accordingly. The varying relationship between architecture and
bridge design has had something to do with this change of heart.

Until comparatively recently bridges were similar in genesis to other types of
architecture; that is, the more important were designed by architects, the less pretentious
by anonymous local craftsmen. This fruitful unity of the structural arts was broken in the
eighteenth century by the invention of engineering as a separate, highly specialized
profession. The split between science and art was not abrupt. In fact, they continued
on generally friendly, mutually sympathetic terms through the early nineteenth century.
Gradually, however, architecture tended to deteriorate into mere decoration, and
architects stewed contentedly in their own precious juices of stylistic revival and eclec-
ticism, divorced from the reality of the great new building problems and the great new
materials that were to solve them.

It has been customary to lament the break between architecture and engineering,
but amateurs of nineteenth-century metal bridges should be grateful that the Battle of
the Styles was not fought over a bridge-head. For the engineers, seeking no justification
in historical precedent, were free to find appropriate expression for the new materials.
And they were free to create new esthetic values through the revealed energy and
the almost miraculous lightness of their gravity-defying spans. Thus the best work of
the engineers was more truly architecture— in the proper sense of that word— than the
nostalgic re-creations and adaptations of those who called themselves architects.

The twentieth century has witnessed a curious reversal of position. Unlike the great
engineers of the preceding century, who saw that bridge construction and bridge
esthetics were inseparable, and felt quite capable of solving both, together, today's
engineer rarely looks beyond his ever more formidable trove of scientific and technical
knowledge. Contemptuous of art, he tends to satisfy himself with mere expediency.
When specially called upon for beauty, he usually, either by himself or with the advice
of a decorator-architect, seeks to embellish his indifferent structure with some kind of
external "styling," thus confusing whatever clean, inherently expressive lines his design
might originally have possessed. This "styling" assumes many guises, but most often
it is an attempt to recall the massiveness of stone construction— sometimes even its
specific shapes, especially that of the time-hallowed arch. It is a strange fact that
lightness is more readily accepted in horizontals than in verticals. Horizontal members
may be distorted into arches, masked with a stone shell, or laden with vulgar ornament,
yet they are rarely deliberately thickened. But it is with real gusto that today's run-of-
the-mill bridge designer sets about the work of dramatizing the solidity of his vertical
supports, particularly if these supports are the abutments of real or apparent arches.
It is worth noting that the architects, who invented the ingenious cosmetics now cherished
by the engineers, are themselves rapidly learning to discard them. Contemporary
architects are assimilating the lesson of nineteenth-century engineering; perhaps con-
temporary engineers would profit from a study of the principles of twentieth-century
architecture.

It would not be fair, however, to blame the faked massiveness of most of today's
bridges entirely upon the engineers and their consulting decorators. Often it is popular
and official taste that is culpable. Even the great Maillart (pages 102-113), when he
was finally allowed to build a bridge in Switzerland's capital city of Berne, was forced
to sacrifice his proposal of a lithe and elegant three-hinged arch to the official demand
for a massive stone-like vault.

Such superficial beautification is far more common in the United States than in
Europe. And what is more, American bridges are actually grossly over-dimensioned
as compared with their European counterparts. That is not wholly the fault of the
American engineer, for he works under a terrible handicap: American materials are too
cheap. Europe, with its historic pattern of relatively expensive material and relatively
cheap labor, has been pushed into extremely economical design and the inventior of

INTRODUCTION 8

new and ever more efficient ways of building. Scarcity of material has also encouraged
good craftsmanship, when one has a single stick of wood one handles it with love
and care.

In the United States conditions are quite the opposite. Competent labor comes high;
and fine spare design, with its demand for careful computation, special steelwork, highly
skilled workmanship and conscientious supervision, is almost prohibitively expensive.
It is much cheaper just to throw in a few more yards of concrete, a few more tons of
steel. But the advantages of economy of material are proportionate to the length of
a span, and enormous spans are feasible only when the "dead load" of the material
itself is reduced to a minimum. This was the special incentive for the magnificent
slenderness of our great suspension bridges— our one important contribution to modern
bridge design.

Almost any American is alert to the airy beauty of these suspension bridges. Curious,
then, that he should distrust lightness in other types of bridge design, and that he
should look to massiveness for his pleasure. Sometimes he tries to justify his prejudice
by claiming that lightly drawn bridges look unrelated to their surroundings. But that
argument has small validity. For when the scale of a bridge seems wrong the fault is
almost always one of brutality. Massive concrete arches, for example, can dwarf and
distort a man-scaled urban or rural scene in most distressing fashion; and even in
grandiose natural terrain, where a heavy structure might seem justified, the contrast
of a delicately membered bridge may be far more delightful. The spidery trusses of
nineteenth-century viaducts are as good a case in point as the tenuous lines of our own
suspension bridges. Far better merely to ask that a bridge disturb its surroundings as
little as possible than to seek an over-literal harmony.

Economy of material cannot alone assure design excellence. Nor is it enough to
add a demand for justice of proportions, for the refinement of structural elements and
the clarification of their relationships. A bridge can be much more than the sum of
these rather negative virtues. It can be the bringing together of material, structure
and form as one thing, one song in space.

Integrity in this sense is inherent, not imposed. It is not a question of paring
down, nor has it to do with mere arrangement or composition. It seems to come only
from the conception of structure as an organism developing according to the law of
its own nature, quickening inert material into life and giving it meaning.

This esthetic ideal is technically substantiated and physically invigorated by the
relatively new idea of structural continuity. When structure is continuous, a bridge is
no longer an assemblage of separately computed, separately functioning items.
Instead, all elements act together, literally fused into a single working shape. It was
through continuity of structure that the plastic nature of reinforced concrete was first
made explicit, in Maillart's bridges. These prophesy a future in which welded steel
and plastic-bonded plywood, like reinforced concrete, will be molded into thin shells,
stiffened by bending. Abandoning line for surface, skeleton for shell, right angles for
curves, and two dimensions for three, a bridge will become, more than ever before,
a single splendid gesture dedicated to the conquest of space.

As the mad Caligula knew when he recklessly launched a bridge into the Mediter-
ranean, a beautiful bridge has a life quite beyond its purely practical functions.

9 INTRODUCTION

STRUCTURAL TYPES

f

There are only three basic types of bridge construction — beam, arch and suspension
cable. Combinations of these are possible, but the bridge that follows one unified,
clearly defined principle of construction is generally more satisfactory in appearance
than the hybrid. Unfortunately for the amateur observer, however, the structural system
of a bridge, even when pure, is not always immediately recognizable, for it depends
less on superficial form than on the manner in which the load is transferred from the
span to the points of support. In other words, an arch is not necessarily an arch.

Beams neither thrust nor pull: they rest. Their load is transmitted vertically to the
supports, and gravity is the only force involved. Easy examples are the stone clapper
bridge (page 12), the primitive log bridge (page 30), the early iron trussed girders
(page 70), and the reinforced concrete overpass pictured on page 1 17.

Everyone knows from experience that a simple beam tends to bend and break at
the middle of its span. This means that the lower part of a beam is subject to pull or
tension, even while the upper portion is being squeezed together or compressed. Beam
construction is therefore appropriate only to materials strong in both tension and com-
pression: stone beams, though feasible, make little sense, as stone lacks tensile strength.

In modern multi-span bridges the beams are often allowed to run continuously over
a number of supports. These continuous beams use their materials so efficiently that
they can be relatively shallow and light. And unlike the simple one-span beam, their
greatest strength is generally required over the intermediate points of support. There-
fore their depth can logically be decreased toward the center of each span, exactly
where extra headroom is often advantageous. The examples illustrated on pages 67 to
69 and 1 18 to 1 19 show that the lower edge of the beam may be brought down to
the support in either a diagonal or a curve; in the latter case the bridge tends to look
like a series of arches, though there is no arch action involved.

A variation of the beam principle is found in the cantilever bridge. A cantilever is
essentially just a beam that projects out beyond its vertical support or supports. Some-
times two symmetrical cantilevered structures are set arm to arm, perhaps forming an
arched opening as in the viaduct at Viaur (page 78). More often the cantilever arms
do not themselves meet, but are connected by a light "suspended span." The classic
example of this highly developed version of cantilever construction is the great Firth of
Forth Bridge (page 77). A similar system of cantilevered beam ends and suspended
spans is advantageous under certain conditions for continuous beam bridges.

The rigid frame (or portal frame, as it is sometimes called) is a special case, for the
beam is monolithic with its supports, and horizontals and verticals form a rigid unit.
Here again, the depth of the span may profitably be reduced toward the center and
the transition effected in a smooth curve. The result may be an arch in appearance,
but it is still a beam in action. An example is the Dry Creek Bridge on page 1 16.

Arches are much more lively than beams, for they are constantly pushing outward
against their supports or abutments. Since the load of an arch is transferred diagonally
rather than vertically to its supports, the planes of contact must be inclined, and the
abutments must be strong enough to meet the powerful thrust of the arch. The arch itself
is squeezed or compressed, therefore well adapted to construction in a material strong

r i

STRUCTURAL TYPES 10

%

V

only in compression: stone, or unreinforced concrete. Depending upon the firmness of
the ground, the arch is either fixed at each end or provided with two or three hinges
to allow for possible movement.

The fixed arch may be of uniform depth throughout, but in its most expressive form it
is shallowest at the crown (the top) and grows deeper as it approaches its abutments.
Fine examples are the wooden footbridge on page 38, the Russian Gulch Bridge (page
89) and Freyssinet's famous through-arch at St. Pierre du Vauvray (page 99). Stone
arches are invariably fixed.

The two-hinged arch is provided with a hinge at each abutment, where its load is
concentrated in a point. Its most dramatic form is the crescent or sickle-shaped arch,
best illustrated by Eiffel's viaduct at Garabit (page 43).

The three-hinged arch allows for movement at the crown as well as at each end.
In unskilled hands it tends to be an awkward bulging shape, for maximum thickness is
required at the quarters; but Maillart, the Swiss engineer, developed it into a thing of
beauty. His variations on the theme are illustrated on pages 1 06 to 111.

Suspension cables are like arches in that their reactions at the abutments are in-
clined, not vertical. But they are reversed arches, for they pull at their abutments rather
than push, and the cables or chains are wholly in tension whereas arch ribs are wholly
in compression. Therefore the supporting members of suspension bridges must be com-
posed of material that is not only flexible, but strong in tension: twisted vines or rope
(page 32), metal chains (page 55) or steel wire (page 58).

11 STRUCTURAL TYPES

STONE

Stone has an importance far beyond its limited use today, for it was in stone that
the building of bridges first became a conscious art, and it is therefore stone that, for
better or worse, has determined many of today's attitudes toward the esthetics of
bridge design.

A slab of stone is an unlikely medium for a horizontal beam, for it is really strong
only in compression whereas a beam must also be strong in tension if it is not to crack
and break at the middle. Yet the beam rather than the arch was the first thought in
stone construction— whether of buildings or of bridges— for it was the easier principle
and familiar to any people who had dealt with wood.

When the Romans gave the arch to western civilization they gave the effective
masonry bridge, for it is in the arch that the compressive strength of stone or brick comes
into its own. The Roman pattern was simple enough: semicircular arches, rarely wider
than 80 feet, supported by thick piers (usually about a third of the span) that took the
lateral thrust of adjacent arches as well as their weight and thereby made each arch
completely independent of its fellows. Piers were protected on the upstream side by
pointed cutwaters and often lightened by small arched openings. The stones were
dressed with utmost precision, and often laid up without mortar.

Most Roman bridges were devoid of ornament other than the strong moldings that
marked the line of the roadway, plus the inherently decorative quality of fine stonework
itself. Some of the urban Italian bridges, however, were embellished with projecting
pillars or pilastered niches as well as elaborate cornices. Like their Renaissance and
post-Renaissance imitations, these were better as buildings than as bridges, for the
extraneous vertical accents tended to disrupt the sense of unity of purpose and continuity
of line that is the very essence of a fine bridge.

Eastleach Martin Bridge over the Leach, England.
Date unknown.

England has numerous slab or "clapper"
bridges, some of which date from pre-Roman times
while others seem to be less than three hundred
years old.

The Greeks developed stone post-and-lintel
construction to a peak of refinement in their temples,
but showed little interest in bridges for they were
a seafaring people and their waterways were their
highways. The boldest stone slab bridges were
built by the Chinese: the Cyclops themselves would
have boasted of the fabulous 70-foot clear spans
of the thirteenth century lo-yang Bridge in Fu-kien.

STONE 12

Pons Augustus (Ponte di Augusto), Rimini, Italy.
20 B. C. Spans of 28 and 23 feet.

Best preserved of the famous Roman bridges in Italy, this is
characteristic in its nobility of proportion and its exemplary
workmanship as well as in such specific features as the semicircular
arches, uneven in number, the thick piers and the strong cornice
with plain, round-topped parapet above.

But only the Ponte Rotto and the Sant' Angelo, both in Rome,
were as elaborately ornamented. The cornice and its supporting
modillions emphasize the continuity of the roadway even while the
pilastered niches with their entablatures and pediments tend to
disrupt it, though time has softened their outlines.

Beloved of Palladia (see page 20), this bridge, through hit
influence, has been imitated in many parts of the world

The diagram illustrates a popular type of Roman bridge, executed
with particular dignity in Spain. The small arches over the angular
cutwaters serve the practical purpose of lightening the massive
piers and allowing additional passage for flood water; they also
provide a counterpoint to the rhythm of the main arches. Beauty
is sought and found in refinement of structure, and there is no
recourse to extraneous ornament.

ROM.TIBERBROCKE, 40 V. ZW.

13 STONE

European society was so disorganized after the fall of the Roman Empire that bridges
must have seemed neither necessary nor desirable, even if the art of their construction
had been understood. The old Roman bridges were ravaged by war and neglect, and
nothing new worth mentioning was built until the twelfth and thirteenth centuries, when
the art was revived by groups of monks, Pontist Friars, who charged themselves with
the assistance of travelers and pilgrims, particularly through the construction of bridges.
The famous bridge at Avignon (below) was their work, also the original London Bridge
(opposite). Later bridges, often heavily fortified, were built not by religious orders, nor
often by feudal lords, but by the increasingly powerful cities.

Medieval bridges were crudely built as compared with their Roman ancestors, and
some have little more than quaintness to recommend them. But by and large, medieval
builders compensated for their technical shortcomings by the fresh, sure intuition with
which they approached their ancient structural system of high-curving barrel vaults and
massive piers. The very rudeness of their stonework is more lively than the exquisitely
dressed, painstakingly coursed masonry of the Romans. And when they achieved the
majestic simplicity of Newby Bridge (page 17) or the famous arch at Lucca (page 25),
their shapes seemed to flower in beauty from some special awareness of the unity of
material, structure and form.

The aspiration toward soaring lightness that was manifest in the great Gothic
cathedrals was reflected in contemporary bridges, to such extent that daring occasionally
exceeded ability, but curiously enough, the basic principle of cathedral construction—
the concentration of arch weight' and arch thrust upon isolated points of support— was
not applied to bridges. Until the advent of Perronet (page 19) stone bridges were to
rely for their strength upon brute mass alone.

Pont d Avignon, over the Rh6ne at Avignon,
France, 12th century. Built by St. Benezet,
leader of the Pontist Friars in France.

This was the first of the great medieval
bridges. Only four of the arches remain, yet
they are sufficient indication of the special
character of the bridge as distinguished
from its Roman precedents. The old
Roman discipline has gone, along with
the fine workmanship, but new values
appear in the flattened curve of the arches
and the off-center accent of the little
chapel that is rooted in one of the great
piers. There is no dominant axis, no
interruption of the rhythmic flow of the
bridge by emphasis upon its center or
its ends.

The apparent casualness of the
composition is deceptive, for here,
conscious or unconscious, is evidence
of a new will to form.

STONE 14

Mil! 'J a k- utlfsmi *?*. tti M b*M <• k J

IvMtWA gEB ggggsaa w www

Old London Bridge. Started in 1176 by Peter Coiechurch, a Benedictine monk who
belonged to one of the famous bridge-building orders of the time; completed
in 1209; replaced in 1824.

This first stone bridge over the Thames was crude indeed as compared with the
work of ancient Rome, or even of contemporary France. The pointed arches, a
medieval invention, exert less outward thrust than semicircular arches, but the
advantage was not exploited. The narrowness of the irregular spans together with
the great breadth of the piers and their even broader cutwaters made the
bridge into an almost impassable dam.

This print of 1594, drawn by one John Norden, bears the following legend:
"There inhabit upon this bridge about 100 householders where also are all kinds of
wares to be bought and sold; the houses are on either side so artificially combined
as the bridge seemeth not only a continuous street but men walk as under a
firm vault or loft."

15 STONE

Great Haywood Bridge, Staffordshire, England.

The medieval bridge pictured here is very similar in design to the one illustrated
on the opposite page. The roadway is so narrow that the triangular niches over
the cutwaters offer useful shelter from oncoming traffic.

The provision of cutwaters on both faces of a bridge was a medieval innovation,
technically preferable to the Roman practice of using cutwaters only on the
upstream side.

STONE 16

Newby Bridge, Lancastershire, near Lake Windermere, England.

One of the most satisfactory of many handsome medieval English bridges,
this is typical in its disregard for the line of the roadway. The parapet is
continuous with the spandrel walls and even follows the sharp diagonals of the
cutwaters, thus making the bridge into a rhythmic alternation of just two
elements— rounded arches and elongated, projecting piers. The lively surface
of the rude stonework is an excellent foil for the boldly defined shapes.

The level of technical achievement is thoroughly unremarkable, yet the expression
of the Roman principle of massive piers and independent arches is as appropriate
as any pattern developed by the Romans themselves.

17 STONE

Santa Trinil6 Bridge, Florence. Designed by Michelangelo Buonarroti; built by
Bartolomeo Ammanati between 1566 and 1569; destroyed in World War II.

With its flattened, subtly curved arches, its exquisite proportions, its air
of certainty and restraint, this was surely the most beautiful and original of all
Renaissance bridges.

STONE 18

m

For many centuries the bridge had been basically Roman. The Gothic cathedral
builders had not cared to apply their revolutionary techniques of masonry construction
—the ribbed vault and the flying buttress— to the humbler problem of the bridge.
Renaissance architects had flattened and refined the semicircular Roman arch, but the
structural principle itself remained unchanged until Perronet, French master builder of
the eighteenth century and one of the world's first professional engineers, happened
on the idea of interdependent arches.

The idea of transferring the thrust of an arch beyond its immediate vertical supports
was not new, for it had been the principle of the Gothic flying buttress; but Perronet
was the first to apply it to bridge design. By using his piers to take only vertical loads,
and letting the thrust of the arches carry over from one to the next until it was met at
the ends by strong abutments, he was able to reduce the thickness of his piers to less
Than half the usual ratio.

^Perronet immediately realized the esthetic consequences of his technical innovation.
The flattened arches now assumed a life of their own, continuous from one abutment
to the other, and quite separate from the slim piers that raised them into the air. His
Neuilly Bridge (below) in particular, with the lean, leaping curves of its arches and
the long unbroken lines of its cornice and parapet, was a classic statement of the idea
of continuous structure. As such, it is interesting to compare with today's continuous
beams of steel or reinforced concrete.

Am/ ,/.■ Xsui/fy.

Bridge over the Seine at Neuilly, near Paris. Built 1768-74; demolished in 1938.
Jean Rodolphe Perronet, engineer. Five 120-foot spans; piers 13 feet thick.

The elliptical arches were splayed out to become segmental arches at the face.

19 STONE

If any one person could be held responsible for the split of engineering from archi-
tecture it would be Palladio, the celebrated mid-sixteenth-century Italian architect. This
he accomplished less by merit than by demerit, for as prime mover in the revival of
ancient Roman glories he was the first and most influential representative of the architect
as reviver and picture-maker rather than as builder. That was not too pernicious as long
as the conditions of living and building remained fairly static, as they did through
the eighteenth century, and as a matter of fact a multitude of fine buildings all over
the western world must be ascribed to the Palladian influence; but it was an attitude
that gave the architect no encouragement to face squarely the new problems and pos-
sibilities of construction that came with the machine. A separate profession had to be
developed to meet the emergency, an engineering profession.

It is therefore interesting to compare the work of Perronet, the first great engineer,
with the bridge designs of Palladio, the man who helped to make him possible. Whereas
Perronet's bridges are bridges, Palladio's are self-contained architectural exercises.
Their function of carrying a continuous roadway over a stream is quite incidental, and
the water beneath seems a fortunate accident, a delightful mirror, rather than their
reason for existence. This is obvious in the Palladian type that is embellished with
shops and arcades, and it is more subtly evident in his adaptation of the Pons Augustus.

Palladio's interest in bridges was not limited to a revival and development of the
Roman masonry style, though that was where he had his great importance. He was
also the first to be concerned with the possibility of wooden trusses, a fact that suggests
that he himself might have welcomed the opportunities of steel and reinforced concrete
without the qualms that many centuries later beset his self-appointed disciples

Bridge design by Palladio, adapted from the ancient Pons Augustus
at Rimini, shown on page 13.

STONE 20

^ivWxK^

Design by Palladio for a stone bridge carrying three separate footways
and six rows of shops.

Pulteney Bridge over the Avon, Bath, England. 18th century.

The Palladian style flourished with exceptional grace in England,
whence it came to us as "Georgian," and the finest Palladian bridges were
built not in Italy but in England. This example differs from others in its
understatement, and seems to carry the city over the water without itself
claiming any excessive importance.

21 STONE

Pont du Gard, the Roman aqueduct at Ntmes, France. 14 A.D. Spans up to 80 feet.

Semicircular arches in three tiers lift the aqueduct 155 feet above the stream.
Mortar wos used only in the top arcade, for the stones of the lower tiers were to
precisely cut and fitted as to require neither mortar nor iron clamps. The colossal
aqueduct is so powerful in outline and proportion that it deserves the immortality
assured by its substantial construction.

WUtaJ*-* ■»•?_ *-l **S *—» — » — >-• ■»•> mn <~ V' *■» w »~- W V—

STONE 22

Londwasser Viaduct for the Rhaetian Railway near Filisur, Canton Grisons, Switzerland.
c. 1904.

Elongated arches and curved plan are brought together in a design of bold
simplicity.

When a stone bridge must be both high and long, the problem is complicated by
the difficulty of spanning great distances with that material. Arches are then
necessarily either ranged in tiers, as in the Pont du Gard, or made uniformly tall and
slim, as in this famous Swiss example, or given varied spans. Since this last
procedure usually means that arches spring from different levels, it rarely leads to
a satisfactory appearance.

23 STONE

The history of bridge building is largely a story of man's willful pursuit of lightness
and his eventual triumph over inert mass. Perronet's interdependent arches were an
important step forward. Another was the development of the open spandrel.

As long as builders knew only the Roman semicircular arch with solid spandrels
(i.e., the walls between supporting vault and deck), a large single-span bridge had to
be extremely high and, unless it were a peaked "camel's back," extremely heavy and
massive. The challenge was thus not only to develop a flatter arch, but to lighten the
weight of the bridge by opening up the spandrel walls.

Spandrel arches meant a gain in intelligibility as well as in lightness, for they threw
emphasis upon the decisive structural importance of the main vault, and differentiated
it from its burden. Today's open-spandrel arches of reinforced concrete are not a
separate phenomenon, but an advanced stage of a development that started in Europe
in the fourteenth century, though the Chinese had solved these problems with great
elegance (see page 26) more than seven hundred years before.

Ceret Bridge over the Tech, France. 1321-39. 147-foot span.

The semicircular arch was in the Roman-medieval tradition, but the small
arched openings in the spandrel walls represented a new thought in European
bridge building.

The original outlines were blurred by repairs and the chapel over the crown,
a characteristic medieval feature, was already ruined when this engraving
was made in 1809.

Cabin John Aqueduct, in Maryland, near
Washington, D. C. 1864. Built by General Meigs.
218-foot span.

The boldest stone arch in the United States
and for forty years the broadest in the world, it
is commendable for its forthright design as well
as for its unusual dimensions. The flat top,
accentuated by strong moldings, is less graceful
over a great single arch than the more usual
curve, but in an aqueduct it was mandatory. No
attempt was made to lighten the massive
stonework above the low-curving arc of
the vault.

This is one of very few important stone
bridges in this country. There are some charming
miniature highway bridges, often dating from
colonial times, and some fine early railroad
viaducts, notably those designed by Benjamin
Latrobe (the Younger) for the B. & O. Railroad;
but American conditions favored bridges
of wood and, later, of metal.

STONE 24

Ponte dello Maddaleno, over the Serchio near Lucca, Italy. 14th century. 120-foot
span.

From the low river banks the road runs steeply up and over the great semi-
circular arch and the smaller side arches, making a "camel's back." Note the lively
asymmetrical composition.

Brig 'a Daon, Alloway, Scotland.

This fine medieval bridge, beloved of Robert Burns, gains in decision through the
crisp molding about its arch ring. Compare with the Swedish arch of reinforced
concrete that is shown on page 91

25 STONE

An-chi Bridge at Choo Chou, Hopei, China. Built by Li Ch'un during the
Sui Dynasty, 590-616 A.D. 1 17-foot span.

This is the oldest open-spandrel bridge in the world. The low-rising arch ring,
a segment of a circle, is brought into sharp relief by the introduction of arched
openings in the spandrel walls. These serve not only to lighten the bridge but to
differentiate cause from effect, i.e., supporting arch from supported roadway.

The structure is phrased with such logic and grace, such acute awareness of its own
nature, that it makes most western bridges seem heavy and inarticulate by contrast.

STONE 26

Doiau-herian Bridge over the Towy near Llandovery, Wales. Designed by William
Edwards (1719-89), who in 1755 built the similar Pontypridd Bridge in South Wales.

Edwards found through sad experience that his broad flat arches were feasible
only when the pressure of the haunches was relieved by spandrel openings. These
cylindrical perforations do nothing to differentiate the basic structural elements of the
bridge, but they contribute a great deal to the beauty of the swelling arch.

27 STONE

Bridge in the Fenway, Boston. 1880-81. Henry Hobson Richardson, architect.

The most beautiful bridges are not necessarily the most daring. Designed by a
famous pioneer of modern architecture, the Fenway bridge is frankly massive. With
a fine eye for texture, pattern and scale, the architect has laid up the great
blocks of stone in freely curving shapes that follow no historical precedent.
If the bridge looks at once medieval and modern; it is because Richardson,
here as in his buildings, used Romanesque masonry as his structural point of
departure, though not as an arbiter of specific form.

STONE 28

Overposs near Eisenberg, Germany, c. 1937
For the Re/chsaufobarin. Karl Schaechterle
and Fritz Leonhardt, chief engineers;
Paul Bonatz, architect.

A bit dry and mechanical, this is
nevertheless one of the most handsome of
recent stone bridges. The spandrel arches
march off over the abutments without
break, and the graceful arcade is a superb
foil for the tapered curves of the main arch
and the shorp horizontal of the deck.

Stone construction is so laborious and time-consuming that it is justly considered an
absurd and costly anachronism in a day when skilled labor comes high, while steel and
reinforced concrete, true machine-age materials, are relatively cheap. Any contemporary
American bridge that purports to be stone should therefore be regarded with suspicion,
for close inspection will usually show that the stone is only a thin layer of deception
applied to a structure of reinforced concrete.

Conditions were different in Nazi Germany. Short of steel and well provided with
highly skilled masons, the Germans found stone bridges a not excessive luxury, and
built many in connection with the impressive network of military highways that was their
Re/chsautobahn. But their reversion to traditional stonework cannot be fully justified
on rational grounds. It must be attributed not only to their shortage of steel, but to
their craving for self-glorification through familiar symbols of power and immortality.

In these German bridges stone was confined to actual vaults and visible surfaces,
while the core was concrete. The stone skin, protective and ornamental, was also in a
sense structural, for it was not an extraneous addition, but integral with the concrete
beneath. Craftsmanship was remarkably good, but the over-all design, though super-
ficially clean, was pallid and affected as compared with the best work of the past.
The old vigor eluded the modern revivalists. Indeed, the German experiment proved
rather conclusively that stone has run its splendid course as a bridge-building material.

29 STONE

WOOD

Kok-su Bridge, China.

A log crib serves as intermediate pier for log spans in this primitive bridge, a
type still built in the American backwoods.

It takes little thought and small skill to bridge a narrow stream with a felled tree,
but the builder in wood— of whatever time or place— has needed considerable ingenuity
to attain spans beyond the length or strength of any single timber. He has met the
challenge in various ways. Sometimes he has shortened his span by projecting the
vertical elements themselves out over the water, using some form of corbeling, bracket-
ing or cantilevering. If he has been familiar with masonry construction, he has often
bound his short timbers together to fashion great arches. Through trial and error he
has also learned to join pieces of wood in triangles, forming strong rigid trusses. And
from ancient times he has used plant material in the special form of twisted vines, bam-
boo splits or hemp rope, as cable for suspension bridges.

Like iron and steel, and unlike stone, wood is strong both in tension and in com-
pression. Rather like iron and steel, and very unlike reinforced concrete, it must be dealt
with as separate pieces. Obviously, then, it is generally suited to the same types of
construction that are propitious in metal, and it is not surprising that the truss should
have had its early development in wood, only later being translated into iron.

WOOD 30

Old wooden truss, Tennessee.

The simplest type of truss is a triangle, from the apex of which a
"king post" is hung to stiffen the horizontal beam at its bending point.

"New Railroad Bridge at Portage, New York." Mid-19th century.
Wooden piecework.

Wandipore Bridge, Bhutan, between India and Tibet.

Great timbers are corbelled out toward each other from massive
abutments and the narrowed interval finally capped with a light beam.
This is the prototype of the modern cantilever bridge with "suspended
span," such as those illustrated on pages 77 and 79.

Thomas Pope, who drew this picture for his book Bridge
Architecture and used it as inspiration for his "flying lever" project,
shown on page 36, said that the main span of the bridge was 112 feet,
and that it was built in the seventeenth century.

Many bridges of this type were built in India and Chino.

Native cable bridge in Colombia, South America. From an old French print.

31 WOOD

Bridge over the Min River, Szechwan Province, China.
Built since 1935. Total length of 1800 feet.

This multi-span footbridge with its twisted ropes of split bamboo follows
ancient structural principles. The Chinese have long been adept with cable bridges,
using bamboo rope or, even before the seventh century, iron chains, and they
have always laid the deck directly on the curving cables.

The little gable roofs poised over the tapered towers protect the heavy timbers
from damage by weather. Through the rhythmic repetition of these spare and
handsome towers the bridge gains extraordinary distinction. One looks forward to
the possibility of continuous-cable multi-span bridges in the contemporary American
terms of flexible steel towers, steel wire coble and suspended level roadway.

WOOD 32

Bridge south of Yunnon-fu, Yunnan Province, China.

Here is perfection of proportion, and subtle variation of the broad flat rectangle
that is the dominant theme of the composition.

33 WOOD

The "Burr-arch" truss.

Below is a half-section of the first
example, a bridge over the Hudson at
Waterford, New York, built in 1804 by
Theodore Burr. Patented in 1817, the
system was used for a majority of our
covered bridges.

At the right is a view through a rela-
tively late version at Rushville, Indiana,
showing the light arch that supplemented
the trusses.

The Town lattice truss.

Patented in 1820 by Ithiel Town, this
truss needed no unusual lumber sizes and
little tedious framing, and so became very
popular. The diagonal web, continuous
over river piers, took more gracefully to
the gable roof than did the arch-and-truss
combinations. This bridge, over the
Connecticut at Orford, New Hampshire,
was destroyed by flood in 1936.

ymMJ/M/M/MJM//mm

*«%fe»

Covered bridges can be beautiful, but they are self-contained, inarticulate, more
like barns than bridges. The excitement of a daring leap is absent, for one cannot
witness the spring. The cover is not for picturesque effect, but to protect structural tim-
bers and their vulnerable joints from rain and snow. The roof also serves as cross-
bracing. Most American wooden bridges were so efficiently covered over that their
external appearance is quite independent of the mystery of their structure. Design
became primarily an exercise in the just proportioning of roofs and walls and openings,
the expression of portals, and the sympathetic relationship of bridge to landscape.

Covered bridges have been popular in many heavily forested countries, particularly
in Switzerland, where some of the boldest in history were built in the eighteenth century
by the famous Grubenmann brothers. But the richest development took place in the
United States during the first half of the last century.

These bridges may look like barns, but their construction is, of course, far more
complicated. Early American examples, like their European forebears, were usually
pragmatic combinations of arches and trusses, for the truss was not yet fully under-
stood and builders found it expedient to lean on the familiar principle of the arch.
Wernwag's Colossus (see page 37), the longest wooden span of all time and possibly
the most beautiful, was a truss-strengthened arch, but most builders used Burr's system
of arch-strengthened trusses. The awkward "Burr-arch" truss was rivalled in popularity
after 1820 by Town's lattice, a true truss in which the arch played no part. Transition
to metal started in the forties with the use of iron rods as tension members in wooden
trusses and interest soon shifted almost entirely to the new material.

WOOD 34

West Hill Bridge, Montgomery, Vermont.

Photo by Edmund H. Royce from The Covered Bridge by H. W.
Congdon. Alfred A. Knopf, N. Y., 1946.

35 WOOD

Let the broad arc the spacious Hudson stride
And span Columbia's rivers far more wide;
Convince the world America begins
To foster Arts, the ancient work of kings.
Stupendous plan! which none before e'er found,
That half an arc should stand upon the ground,
Without support while building, or a rest;
This caus'd the theorist's rage and sceptic's jest.

Thomas Pope's proposal for a "flying lever" bridge and his fash-
ionable couplets are from his Bridge Architecture of 1811, the
first of such treatises to be published in the United States. Pope's
optimistic span was to be of wood. Arched in form, it was yet a

cantilever-beam in principle, with the "flying levers" projected from
great masonry abutments, fitted out on the New York side as
apartments. He made a Vs-inch scale model of half such a bridge
(of 1800-foot span) and according to witnesses the unsupported
arm, 50 feet long, took a 10-ton weight.

Below are structural details of a slightly different version of the
bridge. A fine flourish marks the spring of the intrepid cantilever,
and the chaste pyramidal abutments are an early instance of the
very satisfactory role that the freely interpreted Greek and Egyptian
styles were to play in the bridge architecture of the next few
decades. Pope, who described himself as an architect and landscape
gardener, was much concerned with the appearance of bridges,
and an avowed believer in "mechanical beauty."

WOOD 36

The Colossus, over the Schuylkill ot Fairmount, Philadelphia.
1812; burned in 1838. Louis Wernwag, architect. 340-foot span.

This now seems to have been the greatest span ever achieved
in wood or stone, for Dr. Joseph Killer of Switzerland has recently
demonstrated that the Grubenmanns' Wettingen Bridge of 1764,
long credited as supreme, was 200 rather than 390 feet in length.

There were five parallel laminated arches, each built up of seven

thicknesses of timber, each strengthened by trusses above. Between
these trussed arches ran two carriageways and two footways.

Judging from this view of 1823, the Colossus was as elegant in
appearance as it was bold in structure, Note the rare grace of the
arch, the complementary curve that marks the line of the roadway,
the fine proportions of the windows. The neo-classic portals are
less convincing.

37 WOOD

Private footbridge near Princeton, New Jersey. 1942.
Kenneth Kassler, architect; Kraemer Luks, engineer.

In this clay of drearily unimaginative wooden bridges the freshness and
delicacy of this design are doubly conspicuous.

The tapered curves of the fixed arches are a lucid expression of the transfer
of weight and thrust to the concrete abutments, and the relationship of these flattened
arches to the much gentler curve of the footway is remarkably easy and graceful.

WOOD 38

Temporary bridge for sight-seeing trains
at the Zuka Exhibition, Zurich, Switzerland,
1947. W. Staubli, engineer; F. O. Kalin,
consultant. Three-hinged arch of laminated
wood.

Permanent arched bridges of glued
and laminated wooden strips were built
as long ago as 1907 by Otto Hetzer,
a Swiss, and this exhibition bridge
followed his system of construction. It is
illustrated here less for its intrinsic merit
than for its suggestion of future pos-
sibilities.

The bent and bulging ribs are no
affectation, for their shape is a direct
reflection of the static forces at work in
a three-hinged arch. Rib edges become
jagged as the bonded layers of wood
decrease in number toward the hinges.

The picture shows the bridge under
construction, before the outline of the
arches was confused by the introduction
of a second, lower deck.

A century or so ago wood lost its reputation as a suitable material for fine bridges.
On every count— except, occasionally, that of initial cost— it seemed hopelessly inferior
to the new materials offered by the Machine Age. Where its cheapness was a conclusive
argument in favor of its use, the result was usually just a humble trestle or a bulky,
awkward truss, best relegated to as remote a location as possible.

After this long period of neglect, wood is beginning to regain its lost dignity.
Through the application of science, stimulated by the wartime shortage of steel, wood
suddenly becomes a modern material. New methods of treating timber give it promise
of relative permanence; the new metal ring connectors offer an easy and efficient
substitute for the ancient and laborious practice of framing one timber into another in
complex, vulnerable joints; and the new methods of lamination, of plastic-bonding
short pieces of wood together to form huge monolithic beams and arches, offer
interesting possibilities for new structure and new form.

A revival of the art of wooden bridge building seems due. Wood may never com-
pete seriously with steel or reinforced concrete for long spans, but for small, light, or
semi-permanent bridges its new potentialities for efficiency— and delight— are yet to be
seriously explored.

39 WOOD

METAL ARCH

Thomas Telford's project of 1801 for a "cast-iron bridge, consisting of a single arch
600 feet in the span, and calculated to supply the place of the present London Bridge."

The structural principle is that of the stone arch, for Telford pfanned to use
Paine's system and build up his vault of small pierced blocks; but the transparent
filigree, seemingly without weight or substance, is an imaginative interpretation
of the special nature of the new material.

This elegant arch with its unprecedented span was unfortunately never executed,
not because difficulty was foreseen in its construction, but because of its high
ramped approaches, unwieldy in the city plan, and because of the general
uncertainty of the times.

Coalbrookdale Bridge over the Severn, England. 1775-79. Designed
by Thomas F. Pritchard, architect, for and with Abraham Darby III
and John Wilkinson. 100-foot span.

This first iron arch is still in good condition, though the pressure
of the earth behind the abutments has pointed what was originally
a semicircular arch. The five separate arch ribs were cast in full
halves at a nearby foundry— a construction very different from
that of Paine's invention.

Not until 1836 was an iron bridge built in the United States.

Sunderland Bridge over the Wear,
England. 1793-96. Rowland Burdon,
builder. 236-foot span.

In 1790 Tom Paine had set up a
successful experimental 110-foot arch in
Paddington Green, London, placing it
on exhibition with a shilling entrance
fee, but the Sunderland Bridge was the
first actual example of his ideas, and
incorporated material from his
experimental arch.

The abutments and the curve of the
roadway are singularly awkward, though
the arch is graceful enough. Each rib
was built up of 105 of the cast-iron
panels that are shown in detail at the
foot of the page. Iron hoops filled the
gap between arch and deck.

Component parts of the Sunderland arch
ribs, from a contemporary engraving:

A. Side view of a block, about 5 feet high.

B. End view of a block.

C. One of the wrought-iron bars that
join the blocks to form a rib.

D. One of the screw bolts.

E. Long view of one of the tubes used
to unite the ribs horizontally.

F. End of a tube.

G. Four blocks united to form part of
two adjacent ribs.

I

D
\

METAL ARCH 40

The great era of iron and steel was the nineteenth century, when most of metal's
magic possibilities were explored and the very soul of the material revealed. Today we
perfect and adapt, often with great skill, and occasionally we use our wealth of scientific
knowledge as a basis for invention, but the original creative gusto seems somewhere
to have been lost. Perhaps a material that is young and fresh is most stimulating to
men's imaginations, closest to their hearts.

A full understanding of the new material was not immediate. When iron first
appeared on the European scene as a likely structural medium for bridges, in eighteenth-
century England, the impulse was to treat it like stone. When a material is so new that
its own individual nature is not yet understood, the usual tendency is to handle it in
the same manner as more familiar materials. Some iron chain bridges were built, as
early as 1741 (see page 55), but the other early iron spans were all arches, and
generally assembled like stone vaults of small panels of cast iron (see opposite page).
This type of construction— actually quite sensible in cast iron, a brittle material that
takes best to direct loads— was invented in 1786 by Tom Paine, the extraordinary
American who was later to turn to political philosophy.

At no time, however, was there any imitation of the superficial appearance of
masonry, or any attempt to duplicate its weightiness. Quite the contrary, iron seems
to have been welcomed from the beginning as an honorable material, cbpable of a
new and startling beauty of its own, and the transition from stone to metal, from mass
to line, was accomplished with a minimum of esthetic fumbling. Masonry remained, of
course, as piers, abutments and towers, and until 1850 or so the engineers, with or
without architectural assistance, seem to have well understood the importance of shap-
ing it into bold, clean-surfaced masses as a foil for their spidery ironwork.

The seventies and eighties saw the development of the modern bridge engineer,
product of standardized scientific training. In the same period came the introduction
and acceptance of steel as a material far stronger and more adaptable than iron.
Exploiting new alloys, metal spans have become constantly longer, lighter.

But the gains of the last sixty years have been more quantitative than qualitative.
Could it be that we are too intimidated by our science to preserve the courage of our
intuitions?

41 METAL ARCH

ennui 1:1.1. it in i'.

/"

Craig Ellachie Bridge over the Spey,
Banffshire, Scotland. 1813. Thomas Telford,
engineer. 150-foot span.

More than other engineers, the great
Telford was interested in economy of
material, and the beautiful bridge that
he flung over the Spey is extraordinarily
delicate.

With its trussed arch ribs and its
trussed spandrels above, this is generally
credited as the first modern metal arch.
It is still in use.

Chepstow Bridge over the Wye,
Monmouth County, England, c. 1800.
John Rennie, engineer..

This is a very early and appealing
example of the iron multi-arch bridge.
Like an attenuated spider web the
arches stretch from one shore to the other,
continuous beneath the long sweeping
curve of the roadway and the light
railings. The masonry is kept low, definitely
subordinate. Compare with the Perronet
bridge illustrated on page 19, and con-
trast with the Eads Bridge pictured below.

Eads Bridge over the Mississippi at St.
Louis. 1868-74. James B. Eads, engineer.
Spans of 502, 520 and 502 feet.

Technically the bridge was a great
triumph, for its arches were of record
span and it marked the first use of tubular
structural members, of big pneumatic
pier-caissons, and of extensive steel.

Eads' vigorous design is well illustrated
by this rare early photograph.

The alternation of flat trussed arches
with substantial masonry piers established
an American design formula that is still
with us today, frequently under the most
unlikely of circumstances.

METAL ARCH 42

Viaduct over the Truyere at Garabit, France. 1884. Charles Eiffel, engineer.
Two-hinged arch of 545-foot span.

Like Eiffel's famous Tower in Paris, his Garabit Viaduct is an early triumph
of French engineering. Its great crescent arch asserts the concentration of forces
at the two abutment hinges, and the powerful outlines of the whole are compli-
mented by the lacy trusswork.

Yet the design tends to fall apart into its separate elements— arch, deck and
parading piers. Compare it in this respect with the magnificently single-minded
bridge that was built contemporaneously over the Firth of Forth (page 77).

43 METAL ARCH

Rainbow Bridge, Niagara Falls, New York.
1941. Waddell 8. Hardesty, engineers;
Aymar Embury II, architect. Hingeless
arch of 950-foot span.

Designed as a "set piece" with precise
boundaries and highly mannered detail,
this plate girder arch with slim spandrel
posts is nevertheless one of the finest
American examples of its kind.

Marble Canyon Bridge over the Colorado,
Arizona. 1928. By the Arizona State
Highway Department: L. C. Lashmet,
engineer. Two-hinged braced-spandrel
arch of 616-foot span.

The bridge is less forceful as an arch
and less expressive of its two hinges than
the Garabit Viaduct shown on the
preceding page, or the Bietschtal Viaduct
illustrated below; but it is more expressive
of its material than the plate girder arch
shown above. It fits very comfortably
between the canyon walls, and its extreme
delicacy is particularly welcome in the
giant-scaled landscape.

Bietschtal Viaduct near Brig, Switzerland.
1913. Adolf Herzog, engineer. Two-hinged
arch of 311 -foot span.

The transparent, sharply articulated,
powerfully jabbing shapes carry a convic-
tion that has nothing to do with prettiness.
Regrettable, however, is the choice of
points on the slant of the arch for the
support of the underslung side trusses.

METAL ARCH 44

Belt Parkway Footbridge, Brooklyn, New York. 1939. Designed
by Clarence C. Combs, New York City Parks Department.
A three-hinged plate girder arch.

The gracious curve of the arch is pointed up by the radiating
lines of girder stiffeners and railing posts. Even the faces of the

abutments are inclined at a sympathetic angle. But the problem
of how finally to straighten out these diagonals at the ends of the
bridge finds no very happy solution.

The span looks heavier than its load of pedestrians would seem
to justify.

Plate girder arches with a minimum of trussing
handsomest of all possible types of steel bridge
suspension bridge. They can indeed be extremely
vigor of the best trussed arches and are certainly
of their material. A plate girder arch with metal
Bridge on page 46) is, from a distance, almost
reinforced concrete arch with reinforced concrete
shown on page 90, top).

are today generally considered the
, with the single exception of the
elegant, but they seem to lack the
less eloquent of the special nature
spandrel posts (like the fine West
impossible to differentiate from a
spandrel posts (such as the bridge

45 METAL ARCH

Vdsterbron (West Bridge), Stockholm. See opposite page.

Proposed aluminum orch over the Canimar River near Matanzas, Cuba. Designed in
1946. O. H. Ammann, engineer. 600-foot span.

If this arch is built it will be the first large bridge of aluminum. It will be very
light in weight as compared with a similar bridge of steel. It will olso be very
expensive. Yet neither outlines nor dimensions seem visibly affected by the uncon-
ventional choice of material.

In comparison with the Vdsterbron, the Ammann design has its weaknesses as
well as its merits. The fixed arch curves clear of the horizontal deck girders
with a decision lacking in the Swedish bridge, but the heavy rigid frames over
the abutments are less fortunate, while the parallel incisions that adorn the
concrete are a regrettable pseudo-modern cliche

METAL ARCH 46

Vosterbron (West Bridge), Stockholm, c. 1935. By the Stockholm Harbor Board:
Ernst Nilsson and S. Kasarnowsky, engineers. Fixed arches of 668 and 551 feet.

Slender steel posts carry the deck equably over arches, abutments and side
slopes, and the effect of limitlessness is accentuated by the vertical stiffeners of
the deck girders, repeated over and over until they disappear in the dim distance.

With a refreshing absence of histrionics the angular concrete abutments are
concisely tailored to meet the thrust of the tapered, smoothly welded arch ribs.

47 METAL ARCH

Bayonne Bridge, New York and New Jersey. See opposite page.

METAL ARCH 48

Bayonne Bridge over the Kill van Kull, New York and New Jersey.
1931. By the Port of New York Authority: O. H. Atnmann, chief
engineer. Two-hinged arch of 1652 feet.

The arch is magnificent, worthy of its fame as the longest in the
world, and the fine-spun web of its trussing offers magic perspectives.
But the design fails completely at the abutments, where the thrust
of the giant arch is apparently met only by a light steel framework,
an obvious impossibility that calls for explanation. Actually the
weight and thrust of the arch is safely passed through its hinged
lower ends to the massive piles of concrete at the base of the steel
framework; but the engineers seem to have tried to disguise their

hinged arch as a deep-ended fixed arch, a form hallowed by long
association with stone construction, for they planned granite-faced
towers that would look strong enough to take the non-existent
thrust of the thickened arch ends. The idea of the stone facing was
abandoned just as it was in the George Washington Bridge (page 59),
built at the same time by the same Authority, but in both cases the
indifferent framework of the towers, obviously not designed for
display, remains as testimony to the original intention.

Compare this bridge with an earlier two-hinged trussed arch,
the Garabit Viaduct, shown on page 43.

49 METAL ARCH

Alton Railroad Overpass, Mazonia, Illinois. 1939. By the
Illinois State Highway Department.

Rolled steel sections make an unusually neat, though doubtless
expensive solution to the difficult problem of bracing a
overhead arch.

Bridge over the Connecticut at Orford, New Hampshire. 1937.
By the State of New Hampshire Highway Department. Two-hinged
tied arch of 425-foot span.

This view along the roadway is less prepossessing than the side
view of the same bridge, shown below. The confusion of the bracing
was perhaps avoidable only at prohibitive additional cost, but there
could be no such excuse for the massive stepped-back parapets
at the entrances.

Bridge over the Connecticut at Orford, New Hampshire.

Compare the outlines of this "bowstring" arch with the free
intersection of arch and roadway in the otherwise identical
bridge pictured on the opposite page.

METAL ARCH 50

Bridge over the Connecticut from Chesterfield, New Hampshire, to
Brattleboro, Vermont. 1936 37. By the State of New Hampshire
Highway Department. Two-hinged arch of 425-foot span.

The roodway intersects the arch at a level well chosen to flatter

its curve, but the abutments might be more clearly expressed,

better differentiated from the retaining walls behind.

The bridge is painted the color of young lettuce, a refreshing
change from the usual blue-gray.

The designer of an overhead arch is confronted with all the usual problems of
arched bridges plus two that are peculiar to his task. The first has to do with the
overhead bracing, which must be kept as light and clean as possible. The second
problem concerns the intersection of arch and roadway, which should take place at a
point that will not only provide the requisite clearance above water and the needed
convenience of approach, but will also maintain the visual integrity of both arch and
deck through a just relationship of the two.

The ubiquitous "bowstring" arch, in which the roadway ties together the extreme
ends of the arch ribs, seems to destroy both the character of the arch and the continuity
of the roadway, but any other generalization is dangerous, for a relationship that is
satisfactory in one instance may under other conditions be extremely clumsy.

51 METAL ARCH

Highway bridge over the Vilaine at La Roche-Bernard, France. 1912.
Dayde, engineer. Three-hinged arch of 656-foot span.

With a magnificent disregard for conventional canons of beauty, the lightly
drawn, somewhat ominous arch attains a splendor all its own. Perhaps only a
French engineer would have been capable of the gesture.

Compare this with the suave footbridge illustrated on the opposite page.
There Is no question as to which is the prettier of the two, bu* there is also
little doubt about which is the more vigorous.

METAL ARCH 52

Footbridge over Lake Shore Drive, Chicago. 1940. By the Engineering
Division of the Chicago Park District: Ralph H. Burke, chief engineer.
Three-hinged arch of 187-foot span.

The attenuated arch and the long slow curve of the walk are brought
together with complete felicity, their ever-changing relationship measured by
the steady beat of the light steel sections that serve as hangers and posts.
The latticed overhead bracing is fine-scaled and unobtrusive; but the design
of abutments and approaches is mannered and heavy-handed—not to be
compared in quality with the span itself.

53 METAL ARCH

SUSPENSION CABLE

The arch is a matter of weight, gravity and pressure. As such, it is relatively passive,
earthbound. The suspended cable reverses the arch curve and grows wings. Impatient
of gravity, it achieves strength without apparent mass or weight. Substance seems
transmuted to line, inert matter to naked energy.

Large suspension bridges attain a measure of esthetic value on a simple quantitative
basis, for extremes of lightness, length and height are in themselves sufficient to arouse
emotion. But that initial awe can only be sustained by quality-by a pleasing cable
curve, by a just relationship between the main span and the smaller side spans, between
the portion of the towers above the main deck and the portion below, and very im-
portantly, by appropriate design of the towers themselves.

It is in this last respect that bridges most often fall short of perfection. The transition
from the tower of stone to the flexible tower of steel completed the transition from
mass to line and theoretically banished the last remnant of obeisance to the old
masonry-born concepts of strength as weight, beauty as mass; but the implications
involved in this change of material have not always been either welcomed or under-
stood. Regretting the loss of the easy monumental possibilities of solid masonry and
uncertain of esthetic substitutes appropriate to their new material, tower designers have
tended to relapse into a helpless confusion of structural and ornamental or semi-
ornamental forms. _,, , . , .

Rope bridges have been built from time immemorial (see page 31), and iron chains

were used for the purpose as early as the seventh century in the Orient, by 1741 in

Europe, but it was an American, James Finley, who in 1801 first suspended a level

roadway from his shore-to-shore cables, rather than laying the deck directly upon them,

as had been customary; and a Finley-patent bridge of 1816 may have been the first

to use wire cable rather than iron chains. The United States contributed little more

until the middle of the century, but these inventions were immediately followed up in

Europe. The English held to their chains of linked iron bars, and with them achieved

spans that were sometimes miracles of lightness and grace, and all the more effective

by contrast with the colossal masonry of the supporting towers. The French and the

Swiss particularly interested in wire, developed methods of spinning the cable in

position and in 1834 achieved a record span of 810 feet at Fribourg, Switzerland.

But leadership passed back to the United States in 1848, when Charles Ellet built a

bridge of 1 0 1 0-foot span over the Ohio at Wheeling, West Virginia; and thanks largely

to the genius of the Roeblings, it has remained here ever since.

The great modern suspension bridge is an American phenomenon, encouraged by
the peninsular sites of two major centers-New York and San Francisco. Foreign ex-
amples are relatively few, relatively small, though sometimes very handsome. And it is
the only bridge type in which the United States excels in lightness, for extreme economy
of material, therefore of bridge weight, is the sine qua non of such tremendous spans.
If suspension bridges look lighter than other types of steel bridge, it is because
they are lighter. The supporting cable, being wholly in tension, takes full advantage
of the fact that steel is far more efficiently used in tension than in compression. Thus it is
in the suspension bridge that the nature of steel is most completely realized.

SUSPENSION CABLE 54

Winch Bridge over the Tees, England. 1741. 71-foot span.

In this earliest European chain bridge, the flooring was laid
directly upon the cables. Beneath were diagonal chains that served
as wind bracing.

Menai Straits Bridge, Wales. Built 1819-24; recently widened.
Thomas Telford, engineer. 570-foot span.

The flat curve of the iron bar-chains over the main span is admirable,
but the bluntly tapered towers are undistinguished and the support
of side spans by both iron hangers and masonry arches is
disconcertingly redundant.

55 SUSPENSION CABLE

The Avon Competition.

In 1829 the city fathers of Bristol selected the great Telford, who had
just completed the Menai Bridge, as referee of a competition for the
design of a suspension bridge over the Avon ot Clifton. He judged all
entries unsuitable, including the proposal of I. K. Brunei (a), which he
declared too long in span. Asked to submit his own recommendation,
Telford obliged with the fantasy shown at bottom (d). This was approved
but not executed, and in 1830 another competition was held, this
time with Telford as contestant. The winner (b) was the twenty-four-year
Brunei, and the final result was his Clifton Bridge, illustrated on the
opposite page.

%i

Brunei's early designs for the Clifton Bridge.

a) Above is his fine project in the first competition: inverted cables
below roadway for stiffening; Norman castles as anchorages; no towers.
916-foot span.

b) Below is the design on which work was commenced in 1836.

*

c) This design was submitted in the competition by one C. H. Capper,
"engineer." A favorite motif of the time was the built-in medieval ruin,
here at its most delightfully incongruous.

d) Telford's Gothic design for the Clifton Bridge shows that the early
nineteenth-century engineers were not unaffected by the romantic
fallacies then current among the architects. Even the chains were to be
ornamented with fretwork.

SUSPENSION CABLE 56

Clifton Bridge (above and right) over the Avon, Bristol, England. I. K. Brunei, engineer.
Begun 1836; completed after Brunei's death in 1859. 702-foot span.

This is by far the most beautiful of the early suspension bridges, and through its
great height still one of the most spectacular. The cable threads across the void
in a shallow curve over the thin line of the roadway, then dips straight from the towers
to anchorages in the rocks behind. The towers themselves are magnificent. Their
inspiration is Egyptian, but the masonry is so boldly and freely shaped in response to
function and material as to seem inevitable. Their style is finally not Egypt's,
but their own.

The executed bridge is a vast improvement over Brunei's original scheme, illustrated
on the opposite page (b)s superfluous side hangers have been omitted, side cables
drawn taut, and the towers given a much more vigorously expressive shape.

Since work on the superstructure started only the year following Brunei's death,
and since his son and biographer complained in 1870 that "no attempt has been made
to complete the towers according to Mr. Brunei's architectural designs," it is perhaps
a mistake to give Brunei entire credit for the simplification and refinement of the
towers. But the quotation may refer merely to the omission of the cast-iron plates,
decorated with Egyptian-style figure drawings showing various stages of construction
work on the bridge, that Brunei had hoped to use as sheathing for the masonry.

57 SUSPENSION CABLE

Brooklyn Bridge, New York, 1869-83. Designed by John Roebling;
executed by his son, Washington A. Roebling. 1595-foot span.

In the Brooklyn Bridge, two materials of opposite nature are brought
together in harmony: granite, strong in compression, piled majestically
into the sky; steel wire, strong in tension, spun lightly through space.
Last and noblest of the great stone-towered suspension bridges, this was
also, in its fantastic boldness, its wealth of technical invention and
very particularly, in its use of steel wire, the prototype of the huge
bridges of the 1930s. The 135-foot clearance set the present standard
for bridges over navigable waters, and the diagonal storm-stays,
radiating down from the tower tops, are being reintroduced today as

a precaution against such a disaster as befell the Tacoma Bridge
(page 63). Here storm cables and vertical suspender cables make a
diaphanous web.

Suspension bridges generally look best when side openings are less
than half the length of the main span, and when the cables at mid-span
curve clear of the roadway. The Brooklyn Bridge meets neither condition,
for its span division is 930-1595-930, and its cables drop almost to the
bottom of its double deck. Yet these shortcomings are not particularly
disturbing, perhaps because of the strong curve taken by the
trussed roadway.

SUSPENSION CABLE 58

George Washington Bridge, New York. 1927-31.
For the Port of New York Authority:
O. H. Ammann, chief engineer; Cass Gilbert,
consulting architect. 3500-foot span.

As the river view above testifies, the
George Washington Bridge is remarkable not
only for its size, but for the excellence of its
proportions. Note the short side spans, the
shallow curve of the cables and the almost
incredible thinness of the roadway. The bridge is
unique in its lack of longitudinal stiffening trusses
or girders. It is stabilized instead by its own
great weight, for it is the heaviest single-span
suspension bridge ever built.

Thirsty for an appearance of orthodox
monumentality, the designers built up the steel
framework of their 635-foot towers with the idea
of casing it later in masonry. The mask was
omitted because of popular protest, but the
meaningless arches remain as testimony to the
original intention. Nor was appearance improved
by the addition of a top story over the cable
saddles, absent in the early photograph above.

59 SUSPENSION CABLE

Golden Gate Bridge, San Francisco. 1933-37.
Joseph B. Strauss, chief engineer,-
O. H. Ammann, Leon S. Moisseiff and
Charles Derleth, Jr., consulting engineers;
Irving F. Morrow, consulting architect.
4200-foot span, longest in the world.

The bridge is fortunate in its colossal
dimensions, its permanent coat of orange
paint, and its spectacular surroundings, but
the quality of its design is not commensurate
with its size. The towers, looming 746 feet
above water, are capricious in outline and
detail, and take poorly to the relatively
low placed, trussed roadway and the
extraordinarily deep curve of the cables.

Chute du BrOle Bridge over the Gatineau
River, Province of Quebec, Canada. 1938.
Designed by the Dominion Bridge Company,
Limited. 300-foot span.

Quiet, graceful towers and clean,
purposeful lines make this probably the
finest small suspension bridge in the
Americas. The relatively deep stiffening
girders art in the European tradition;
compare with the Rodenkirchen bridge
shown on page 62.

Bronx-Whitestone Bridge, New York. 1939. For the Triborough Bridge
Authority: O. H. Ammann, chief engineer,- Allston Dana, engineer
of design; Aymar Embury II, architect. 2300-foot span.

In this first use of shallow plate girders as roadway stiffeners
rather than the customary deep trusses, the ratio of girder depth to span
was a mere 1/210. Since the Tacoma failure (page 63), the 1 1-foot
plate girders have been reinforced above by trusses to a total depth of
25 feet, entailing complete loss of the original proportions. Diagonal
storm stays have also been added, run from tower tops to roadway
as in the Brooklyn Bridge.

The picture shows the bridge in its original condition, with a
fine-spun elegance of outline and detail that was unique among modern
suspension bridges, although the big and little arches of the tapered
377-foot rigid-frame towers are obviously something of an affectation.

SUSPENSION CABLE 60

61 SUSPENSION CABLE

Rodenkirchen Bridge over the Rhine, Cologne, Germony. 1938-41; destroyed in

World War II. For the Reichsautobohn- Karl Schaechterle, Fritz Leonhardt and A. Klonne,

engineers; Paul Bonatz, architect. 1244-foot span.

This largest European suspension bridge was no miracle of lightness as compared
with American achievements. Its beauty was wholly a matter of terse, highly
articulate structure and exquisite proportions. There was not one empty gesture or
superfluous word, and each smallest part was dignified by its coherent relationship
to the whole.

The low clearance permitted by the Rhine's small-scale shipping activities was
obviously a great advantage to the designers of this bridge.

SUSPENSION CABLE 62

Tocoma Narrows Bridge, Washington. 1940; ruptured by wind
four months after completion. 2800-foot span.

This handsome bridge was modeled after the Bronx-Whitestone,
but was longer in span and measured only 39 feet between cables.
The roadway was stiffened only by 8-foot-deep plate girders, a
ratio of girder depth to span (1/350) that has been exceeded only
by the George Washington Bridge, where weight gives stability.

The distinguished committee that investigated the failure
reported that "excessive vertical and torsional oscillations were
made possible by the extraordinary degree of flexibility of the
structure and its relatively small capacity to absorb dynamic forces."
Many of the early suspension bridges suffered a similar fate.

The failure of the Tacoma Bridge was a shock and a challenge to American bridge
engineers. The immediate response was unfortunate, for apprehensive engineers and
public officials hurriedly retreated to the apparent safety of deep, ungainly stiffening
trusses— the obvious, orthodox antidote to excessive flexibility in suspension bridges.

Yet the disaster may finally prove beneficial, for it has inspired a wealth of con-
scientious research and fresh creative thought, some of which now begins to bear fruit.
Most of the new proposals have to do with ways in which proper aerodynamic design
of the roadway— usually through some type of streamlining coupled with a system of
road vents for free wind passage— will permit retention of its shallowness and flexibility.
But the boldest proposal suggests a radical change in both the structure and the form
of suspension bridges, for it discards the principle of flexibility. Instead, the wire hangers
become the diagonal members of a cable truss of extraordinary lightness and stiffness.

63 SUSPENSION CABLE

METAL BEAM

Britonnio Tubular Bridge for the Che$fer and Holyhead Railway,

over the Menai Straits, Wales. 1846-50. Robert Stephenson, engineer;

Francis Thompson, architect. Spans of 230, 460, 460 and 230 feet.

Trains run through twin wrought-iron beams, laid side by side,
each a continuous rectangular tube of 1511 feet. The marble-faced
towers, the central one of which is 230 feet high, were conceived as
supports for auxiliary chains, but the hollow girders proved so strong in
themselves that the cables were omitted.

This was the first great assertion of the flat beam in modern
bridge building. The construction was revolutionary, but no attempt was

made to recall more familiar structural forms, and no compromise was
allowed to blur the decisive relationship of horizontal and vertical,
metal and masonry. Even the Greco-Egyptian overtones of the towers
seem unaffected and curiously harmonious; compare with Brunei's
masterpiece at Clifton, shown on page 57.

This seems to have been one of the first major examples of the
successful collaboration of a bridge engineer and an architect.
Thompson was also the designer of numerous railway stations on
the same line.

A standard design for underpasses below the German Autobahn, c. 1937.
Karl Schaechterle and Fritz Leonhardt, chief engineers; Paul Bonatz, architect.

The plate girder bridge achieves elegance through refinement of structure.
Vertical itiffeners divide the girder into flat rectangles similar in proportion to the
opening itself, and the continuity of the roadway is stressed by light unaccented
railings, and by the prolongation of the cantilevered sidewalk slab as a coping
over the retaining walls.

Note that the seat of the girder is visible at either end, sharply distinguished
from the retaining walls, and that the banks have been held bock to allow the
main lines of the bridge to come free.

METAL BEAM 64

A plate girder bridge is immobile and a bit dry as compared with an arch or a sus-
pension bridge, and in long heavy spans it is likely to seem gross as compared with a
fine-membered truss. But it is a good simple elementary form, orderly and restful, and
at its best— shallow, cleanly drawn, crisply detailed— it is not only pleasantly unobtrusive
but notably elegant.

It was not until about 1830, when the principles of arch and cable were already
highly developed in metal, though the truss was still pragmatic, still wooden, that
George Stephenson, a famous English engineer, first thought of building one of his
railway bridges of flat solid-walled iron girders. The great Britannia Bridge, shown oppo-
site, was the work of his son Robert. Its tubular structure is of course rather different
from a plain plate girder, but the esthetic problems involved in its design were very
similar. In spite of this illustrious ancestor the plate girder was long dismissed as a
humbly utilitarian kind of construction, useful enough for out-of-the-way railroad via-
ducts but definitely unworthy of creative attention. Only in the last fifteen years or so,
when its economy for small spans has made the steel plate girder universally popular
for highway bridges and overpasses, have its potentialities for beauty begun to be
recognized along with its practical advantages.

The girders need not invariably have a straight lower edge. When they are not
divided into separate spans but made continuous over a number of openings, an eco-
nomical procedure, then greatest strength is usually needed to meet stresses and strains
concentrated over the vertical supports. The girders may merely be thickened at these
points, and their under-edges kept flat, as in the German example pictured on the
next page; or the extra strength may be provided through increased girder depth over
the piers. In this latter case, the transition may be effected with technical propriety either
by smooth curves or by sharp diagonals; the first look well under an up-curving roadway,
the second beneath a flat deck. In either event the girder is logically brought straight
to its end-pier or abutment, and thus differentiated from the true arch construction that
it fortuitously resembles. Flowing curves or sharp bends are also in order in the rigid
frame, and again the temptation to imitate true arches must be checked in the interest
of esthetic integrity.

Because of its lack of structural drama the plate girder more than any other bridge
type depends for success upon justice of proportions and perfection of detail. Sidewalks,
railings and abutments assume decisive importance, and the quality of the whole is very
much affected by the design of the piers: by their spacing, which determines the pro-
portions of openings; and by their shaping, preferably as solid, quiet slabs of masonry,
or as light, expressively contoured rigid frames of steel or reinforced concrete.

65 METAL BEAM

Bridge over the Freiberg Creek at Siebenlehen, Germany. 1938.

For the Autobahn.- Karl Schaechterle and Fritz Leonhardt, chief engineers;

Paul Bonatz, architect.

The colossal masonry piers, flat-sided and slightly tapered, are
capped with rollers to allow for movement of the continuous girders in
response to temperature changes. Vertical stiffeners divide the girders
into up-ended rectangles, and the interval between stiffeners is the unit
that determines the location of sidewalk brackets and railing supports.

The engineers of the Nazi bureaucracy were at their best when
dealing with the sober problem of the plate girder and achieved some
extraordinarily handsome solutions, of which this is one of the finest.
Incidentally, the discussion of the design of plate girder bridges is
a particularly valuable chapter in the excellent book on bridge esthetics
written by this same Schaechterle and Leonhardt: Die Gestaltung der
Bruefce (The Design of Bridges), published in Berlin in 1937.

METAL BEAM 66

Birdsong Creek Bridge, near Camden, Tennessee. 1942. By the
Tennessee Valley Authority. Main span of 103 feet.

Like most TVA bridges, this continuous plate girder is not
remarkably light in appearance, but most remarkably neat. Dominant
are the attenuated, well-differentiated horizontals of the steel
girder, the projected concrete coping, the low concrete parapet and
the metal railings.

This clarity is absent at either end, for girder seat, retaining walls
and parapet are brought together as one inarticulate mass of
concrete. The limits of the bridge are so sharply defined by these
terminal accents as to threaten one's sense of the roadway as
continuous.

Bridge over Fontana Reservoir for the Stecoah-Bryson City Road,
North Carolina. 1944. By the Tennessee Valley Authority.
Spans of 189, 228 and 189 feet.

Instead of simply thickening the continuous girders over the
piers to provide the extra strength needed at those points, the
designers have chosen to increase the depth of the girders, bringing
them down in straight diagonal lines that look very well under
the flat road-deck. The shadow cast by the cantilevered sidewalk
makes the bridge seem unusually lively and three-dimensional.

Now that Fontana Dam is completed, the tall piers are partially
submerged. The tilted roadway is not a photographic distortion,
for one end of the bridge is actually considerably higher than
the other.

Gowanus Elevated Parkway, Brooklyn, New York. 1941.
By the Triborough Bridge Authority.

Proud symbol of a new age, the highway cuts above its dreary
surroundings, its long slim legs withdrawn from chaos. These
rigid-frame supports would be better without their weakly drawn,
arbitrary arches, but otherwise they are remarkably clean
and powerful.

Seen from beneath, the structure of the boldly cantilevered
roadway is very expressive, inherently ornamental as it tapers up
and out from the longitudinal girders.

67 METAL BEAM

Old Alfred Road Overpass, on the Maine State Turnpike, near
Biddeford, Maine. 1947. By the Maine Turnpike Authority.
Howard, Needles, Tammen and Bergendoff, engineers.
Continuous girder with spans of 36, 58, 58 and 36 feet.

American engineers have only recently begun to concern
themselves seriously with the appropriate design of plate
girders, but their products are already noticeably cleaner
and more agreeable than they were ten years ago.

Here the deck is carried by the cantilevered arms of
reinforced concrete piers. The piers themselves are set askew
to parallel the divided highway beneath and thus to offer
it a minimum of interference, physically and psychologically.

Saco River Bridge, on the Maine State Turnpike, near Saco,
Maine. 1947. By the Maine Turnpike Authority: Howard,
Needles, Tammen and Bergendoff, engineers. Continuous
girder with two spans of 90 feet and four spans of 110 feet.

The bridge is run over the river in two separate halves
to carry the divided highway of the Turnpike.

Brackets project from the girder wall to support coping and
railing. Although the shaping of these projections seems
willfully labored on close inspection, they make a merry
effect as they run the length of the bridge.

North Chickamaugo Creek Bridge, near Chattanooga, Tennessee. 1940.

By the Tennessee Valley Authority. Continuous girder with spans of 52, 94 and 52 feet;

bridge width of 30 feet.

Here the TVA engineers and architects have produced one of the finest small
bridges in the United States, Bold horizontals and diagonals are accentuated by the
regularly repeated verticals of the girder stiffeners. Perhaps it is this uncompromising
decision of line that makes the bridge more exciting than the softer, sweeter
design shown at the foot of the opposite page.

Valley River Footbridge, Murphy, North Carolina. 1939. By the Tennessee Valley
Authority. Spans of 52, 78 and 52 feet.

The continuous girder curves down at the intermediate piers, harmonizing with the
much gentler curve of the walkway itself. The flat run to the abutments adds vigor
to grace and successfully cancels the incidental resemblance to arch construction.
Rivet heads pattern the surface of the slender girder, but the usual vertical stiffeners
are absent.

Proportions are excellent. Only the narrow little shelves that support the girder
ends are awkward; better to suppress them completely, as in the bridge pictured
above, or better still, give them more distinct expression, as in the German
underpass pictured on page 65.

69 METAL BEAM

Bridge over the Rio Mqlleco, Chile. 1886-89.

The spidery mesh of girders and towers is unmistakably of the last
century, and probably the work of a French engineer

The bridge is so delicate, so transparent, that it threads over the
valley without seeming to disturb it.

Old iron truss, Tennessee.

When bridges such as this are replaced by a type of truss better
suited to modern traffic, the new bridge rarely offers compensation for
the loss of finely etched lines.

Here the main span is a Pratt-type truss, in which diogonols, pure
tension members, were always iron rods, while verticals were either
of wood or of iron. Invented in the United States in 184^, this was one
of the earliest of scientifically designed trusses.

METAL BEAM 70

Pit River Bridge, over Shasta Reservoir, near Redding, California. 1941,
Designed by the engineers of the U. S. Bureau of Reclamation.

This double-deck bridge is one of the handsomest trusses in this
country. Before the development of the theory of the continuous beam,

the unequal spans would have called for girders of varying depths.

There seems to be no good reason why run-of-the-mill overhead
trusses cannot be comparably quiet and horizontal in design.

Nineteenth-century trusses had a gossamer quality that is rare today. Modern trusses
must in actuality be heavier to take today's heavier loads, but their bulkiness must partly
be ascribed to the use of relatively few, relatively large truss members, whereas nine-
teenth-century engineers preferred a close web of many light members, airy in appear-
ance yet well defined in space as a semi-transparent geometric plane. Piers too, espe-
cially in France, were often of lacy trusswork.

The overhead or "through" truss is economical, therefore prevalent, but pleasant
solutions are almost non-existent. The popularity of the lumpy hump-backed version is
particularly regrettable, for an uneven upper edge generally looks nervous and clumsy,
and complicates even further the complicated problem of overhead bracing. Never-
theless, the sky line of a fully developed cantilever truss with arched "suspended span,"
such as those shown on page 77 and 79, has an expressive vigor that can be quite
magnificent.

71 METAL BEAM

Bridge over the Rhine at Neuwied, Germany,
c. 1934; destroyed in World War II.
Karl Schaechterle and Fritz Leonhardt,
chief engineers.

This truss was dignified by the care for
outline and detail that is normally reserved for
more pretentious structures.

All truss members were inclined at an angle
of 63 degrees. This absence of verticals made
for an unusually coherent pattern, intelligible
from every viewpoint. The smooth rigid-frame
portals were very neat, also the lattice
bracing above the roadway.

Goethals Bridge over the Arthur Kill at
Elizabeth, New Jersey. 1928. For the Port of
New York Authority: Waddell & Hardesty,
engineers.

The internal confusion is typical of
an overhead truss with an uneven upper edge.
Compare with the orderly German bridge
illustrated above.

*

Bridge over the Sitter between Haggen and Stein, Switzerland. 1937.
Rudolf Dick, engineer. Continuous truss with spans up to 228 feet.

Proportions are fantastic, for the tallest pier Is 276 feet high, while the total width
of the single-lane roadway and the two sidewalks is only eleven feet. The deck
and its supporting trusses bulge"out over two of the main piers to allow automobiles
to pass.

The extraordinary lightness and laciness of the trussing is pleasantly reminiscent
of nineteenth-century French practice, but the sweeping high-crossed lines of the
lean, tapered legs make a shape definitely of our own day.

METAL BEAM 72

73 METAL BEAM

, ■ '-jgm v..v

Bailey bridges of World War II. Invented by Donald C. Bailey of the
British Ministry of Supply. Spans up to 240 feet.

Bridges are assembled of prefabricated, interchangeable trussed
panels, each ten feet long and designed for handling by six men. Panels
are pinned together on the ground, then shoved out over the water
on rollers. Each truss can be built up to a maximum of three panels in
height, three in thickness.

Through the regularity of its openwork pattern and its development
in bold horizontals, the Bailey truss lends itself to handsome effects.
The trusses pictured above replace wrecked arches of an old Italian
bridge. The short top tier, placed where maximum strength is needed
in a simple (as opposed to a continuous) beam, gives the bridge
an unexpectedly bold and lively shape.

Among military bridges, however, Caesar's pile-and-trestle type was
relatively as ingenious, and Xerxes' Hellespont bridge, with hundreds
of high-prowed triremes and penteconters serving as pontoons,
must have been far more spectacular.

METAL BEAM 74

Lift bridge, Japan, c. 1933. 69-foot span.

Counterweights within the tower legs regulate a movable span
distinguished by unusual smoothness of shape and surface. A series
of rigid frames without diagonals, this is known as a Vierendeel
truss in honor of its inventor and chief promulgator, the late
Professor Arthur Vierendeel of Belgium.

This Japanese adaptation is shown in preference to any of the
hundred-odd examples built since 1896 in Belgium and the Belgiar
Congo because it best suggests the very considerable esthetic

potentialities of the unique construction. The early Belgian Vierendeels
were riveted rather than welded, therefore lack the smoothly flowing
lines and planes of this all-welded structure; and the more recent Belgian
examples, though welded, have a full-curved upper edge that gives
them the appearance of arches and is much less forceful than the
straight-ended truss shown here. Moreover, the roadway in Belgian
practice is normally placed high in the truss, confusing its outlines.

The few Vierendeel trusses that have been built in the United States
are so heavily dimensioned as to appear brutal.

Engineers and architects are beginning to explore the new opportunities for struc-
ture and shape that are offered by electric arc welding. If the visible effect upon
bridge design has thus far been negligible, it is because welding has been used to
lighten and smooth familiar bridge forms rather than as basis for the creation of new
structural shapes.

By eliminating old-fashioned rivets and the angular joints that accompany them,
welding allows a one-piece homogeneous structure with a continuous flow of forces
from one part to the next. The relative efficiency of this assembly method is illus-
trated by the Vierendeel truss pictured above, which weighs a fifth less than a con-
ventional riveted truss of similar design.

Welding implies continuity of structure. When this potentiality is more fully realized
we shall have steel bridges unlike any we have known. Steel will be formed into thin
sheets, stiffened by bending, and these light shell-like structures will have the strength
to span great distances in one smooth leap. They will look more like reinforced con-
crete bridges of advanced design than like the steel bridges of today, for they will
be based on the same principle of structural continuity.

The vertebrate, now supreme, will be challenged by the crustacean.

75 METAL BEAM

Double-leaf bascule for the Canadian Pacific Railroad,
between Sault Ste. Marie, Ontario, and
Sault Ste. Marie, Michigan. 336-foot span.

The great steel arms with their massive counter-
weights are so nakedly expressive of their capability
of sudden movement that the bridge seems like
some giant insect.

Pulaski Skyway over the Hackensack and Passaic
Rivers, Hudson County, New Jersey. 1932. For
the State of New Jersey: Jacob L. Bauer, chief
engineer. Main span of 550 feet.

The Skyway undulates high over the Jersey
meadows, its continuous trusses swung overhead
where extra clearance is needed beneath. Discounting
the whimsy of its pier design, it is more plausible
than most serpentine trusses for it makes no effort
to appear as an arch.

Bridge over the Rur at Duren, Germany. Built 1930;
destroyed in World War II. 256-foot span.

This was a bridge without right angles. The triangle
of the opening was repeated in the arrangement of
truss members and in the design of railing supports.
A curiosity anticipated by Brunei in his Chepstow
Bridge of 1852, shown on page 81, this triangular truss
was recommended by prominent German engineers
as economical for medium-length spans.

METAL BEAM 76

Firth of Forth Bridge, Scotland. 1883-89. John Fowler and
Benjamin Baker, engineers. Two 1700-foot main span*.

The idea of the cantilever is ancient in the Orient (see page 31), and
the German invention of the modern metal cantilever truss dates back
to 1867, but the fabulous Forth bridge was the first major example.
For thirty years its spans were the longest in the entire world.

Few other bridges approach it in dramatic content. The great
tapered towers with their outstretched cantilever arms have a splendid
sweeping fullness, and their assembly of large tubular members makes

their structure extraordinarily intelligible. Small truss members would
have been confusing in these irregular shapes. Every element of
the design is clearly articulated, from the four separate circular piers
under each tower to the arched "suspended spans" that join the
tips of the cantilevers. Even the difficult juncture with the latticed
side-spans is accomplished without fumbling.

The Forth Bridge is not conventionally pretty or graceful, but there is
a deep emotional satisfaction in its powerful lines.

77 METAL BEAM

Viaduct at Viaur, France. Before 1903. Designed by
the Societe e/e Construction des Safignolfes.
722-foot span.

The two great cantilevers meet at the center
without the introduction of a "suspended span,"
making an arched opening.

Twin Falls—Jerome Bridge, Arizona. Before 1927.

Here the "suspended span" is the parallel-edged
section at the center.

The thinly etched lines of the trusswork contrast
very happily with the massive walls of the canyon.

&

is.NSS'EINBSaSir -

Grand Giaize River Bridge, Missouri. 1930.
Sverdrup & Parcel, consulting engineers.

The fully developed cantilever truss is startling in
this underslung version. The piers ore now almost
entirely submerged.

Wuifi.

METAL BEAM 78

Cooper River Bridge, Charleston, South Carolina. 1920.
Waddell & Hardesty, engineers. Two separate cantilever trusses,
the larger with a main span of 1050 feet.

This is certainly not the most beautiful bridge in the world. But
perhaps if is the most spectacular, for here is a highway recklessly
launched into the sky. Steep approaches, stupendous height, extremely
narrow width and a sharp curve at the dip conspire to excite and
alarm the motorist, even while his changing perspective of the second

span gives him multiple awareness of the structure that is hurling
him through space. Perhaps all bridges should be bent at the middle
so that no one might traverse them unaware.

From any viewpoint the long unbroken thread of road somehow
manages to tie together the disparate means of support, and the
looming batlike shapes of the cantilever trusses dominate the skyline for
miles around. This is a bridge for the collector of bridges.

79 METAL BEAM

Bridge over the North Elbe at Hamburg, Germany, e.
Three 330-foot spans.

1882.

Sagamore Bridge over the Cape Cod Canal at Bourne, Massachusetts. 1935
Fay, Spofford & Thorndike, engineers,

A more graceful version of the ambiguous serpentine motif.

The bridges illustrated on these and the two following pages seem to have no one
clearly dominant structural idea: in the same span the beam may, without much show
of favoritism, be combined with cable or with an arch, or arch and cable may be used
together, or even all three at once. Yet all of them are finally better classified as beams
than as anything else, for even the most complicated struggle of forces comes, in the
end, to a neutral deadlock, with little or nothing in the way of external push or pull.

Since some of these hybrids are very good-looking, particularly those shown on
pages 82 and 83, one cannot say that mixed structure is in itself evil. But when the play
of forces is so equivocal that the role of the various members is visually unintelligible,
and the different parts seem mutually contradictory, then the design loses conviction
and the bridge is more curious than beautiful.

METAL BEAM 80

Bridge over the Wye at Chepstow,
England. 1852. I. K. Brunei, engineer
Main span of 300 feet with three
100-foot side spans.

Each of the two tracks is separately
supported, its girders stiffened in the
long jump by bar-chains hung from
either end of on iron tube, and pro-
vided with light vertical and diagonal
stiffeners. The tube resists the pull of
the chains, and the final result is a
truss with a triangular section.

Royal Albert Viaduct over the Tamar at Saltash, England. J 859.
I. K. Brunei, engineer. Two 455-foot spans.

The elliptical iron tubes, 16 feet wide by 9 feet high, act as arches,
and the bar-chains absorb their thrust, perhaps also provide support,
but it is all very mysterious. With its extreme height and extreme
narrowness, accentuated by the slim verticality of the masonry,
this Victorian grotesque manages considerable appeal.

81 METAL BEAM

Pont Transbordeur, Marseilles, France. 1905. Arnodin, engineer.
787-foot span.

The French flair for boldness and lightness in metal construction is
evident in the Transbordeur, which at first sight is a mysterious
arrangement of lines in space-without substance or apparent function.
Actually it is not a bridge but a support for an aerial ferry that travel*

from one bank to the other, suspended just a few yards over the
harbor waters.

From the high towers hang two cantilever beams, joined by a trussed
"suspended span" and anchored firmly to the ground at their far ends.
Compare with the cantilevers shown on pages 77 and 79.

METAL BEAM 82

Bridge over the Rhine between Cologne and Mulheim, Germany. 1929;
destroyed in World War II. Karl Mohringer, engineer. 1033-foot span.
The Mulheim bridge was the finest example of the self-anchored
suspension bridge, a type of beam-and-cable construction that is more
popular in Europe than the pure suspension bridge. The usual external
anchorages are not needed, for the wire cables are attached at either end
to the stiff plate girders of the roadway itself. Since these girders must
absorb the pull of the cables and also provide much of the actual
carrying strength of the bridge, they are rather substantially dimensioned
for American taste. But once the structural premises are accepted, it
must be admitted that the bridge was very skillfully designed:
proportions were excellent and the towers, hinged at the base for
flexibility, were unusually clean in outline.

St. Georges Bridge over the Chesapeake & Delaware Canal, Delaware.
1941. Waddell & Hardesty, engineers; Aymar Embury II, consulting
architect. 540-foot span.

Here the beam ties together the ends of the arch, removing the need
for abutments and thus performing a function similar to that of the beam
which anchors the cable ends of the German bridge pictured above.
In both cases the roadway girders dominate the design, and the
arch or cable is rightly subordinate.

Appearance has been considered with a care not usually lavished on
American steel bridges. Note the neat K-bracing of the orch ribs and,
even more important, the continuation of the lines of the approach girders
in the girders of the main span. This sense of continuity, lacking which
a bowstring arch looks clumsy and lifeless, would have been enhanced
if it had been possible also to extend the detail of the approach
airders— their vertical stiffeners and projecting sidewalks.

Bridge for the Renault Factory, over an arm of the Seine at Billancourt,
France, c. 1932. Etablissements Dayde, engineers.

The two suspended cantilever beams, separated by a "suspended
span," make a construction readily comparable with that of the
Marseilles Transbordeur, but that is the end of any resemblance.

The bridge is beautifully balanced in design, though one might wish
that the suave plate girders had been strong enough to do the job
by themselves, without the assistance of the bar-chains.

83 METAL BEAM

REINFORCED CONCRETE

Composite of steel and masonry, reinforced concrete is often treated as a cheap
substitute for one or the other of these ingredients. Its own separate character comes
out only in the hands of an understanding and sympathetic designer, but then it can
emerge in shapes of rare beauty and distinction.

Plain poured concrete, an ancient concoction of cement and water, sand and
gravel, hardens in molds to become artificial stone. Like natural stone, it is strong only
in compression, therefore suited only to the construction of massive piers and arches.
Not until 1875 or so was it discovered that this man-made masonry might be given
strength in tension through the incorporation of embedded rods of iron or steel. Thus
was born a new and scientific material, reinforced concrete, with the compressive
strength of stone, the tensile strength of steel, plus a plastic quality that is entirely
its own and most appropriately expressed in a fluid continuity of structure and line.

Steel-reinforced concrete is a patient material, all too tolerant of torture. The
plasticity that is its great advantage is also a weakness, for it permits all kinds of gross
indignities. Illustrated here are some of the ways in which it has been abused through
the years; the imitations of stone construction are as patently absurd as the built-in
stalactites shown above; the clumsy truss suggests that structural forms well suited to
assembly from lengths of timber or steel may be foreign to reinforced concrete, and
the elaborate foolishness of the pseudo-modern Connecticut underpass is obvious as
such. Compare these dismal structures with the beautiful bridges on the two following
pages, both completed in 1905, proof that even at that early date there were two
engineers— Francois Hennebique of France and Robert Maillart of Switzerland— who
were successful in creating structural shapes eloquent of the unique powers and prop-
erties of the wonderful new material.

Good spare construction is not easy in reinforced concrete. At every stage it requires
a great deal of skill, care and sensitivity— from the workmen who make the forms, place
the reinforcing steel and mix the concrete as well as from those responsible for design
and supervision. Small wonder then that the best work has been done in Europe, where
thrifty use of material has long been essential and where loving craftsmanship is still
something of a live tradition.

REINFORCED CONCRETE 84

MISUSE OF REINFORCED CONCRETE'.

Alvord Lake Bridge, Golden Gate Park,
San Francisco. 1889. 20-foot span.

This first reinforced concrete bridge in the
United States is still standing. It is less remarkable
for its imitation of rusticated stonework than
for its custom-made stalactites.

Arlington Memorial Bridge, Washington, D. C.
1932. John L. Nagle, engineer; McKim, Meade 4
White, architects.

The bridge is designed in Washington's usual
pompous neo-classic manner. Its open-spandrel
arches of reinforced concrete are faced with
granite slabs in faithful imitation of solid stone
vaults, and its central draw-span of steel is
painted and decorated in faithful imitation of the
aforesaid faithful imitations.

Ridge Road Bridge, Wethersfield, Connecticut,
c. 1938. By Ihe Connecticut State Highway
Department.

This imitation in reinforced concrete of a
medieval stone bridge, such as that illustrated on
page 17, is as inept as it is absurd. Note the
use of pointed cutwaters to divide the traffic lanes.

Reinforced concrete is molded to form a lengthy
truss, gross indeed as compared with the ordinary
steel truss visible at the far left of the picture.

Merritt Parkway Underpass at Stamford, Connec-
ticut, c. 1937. By the Connecticut State
Highway Department.

This rigid frame of reinforced concrete apes no
historical precedent. Its vulgar ornament is
peculiar to our times and easy of achievement in
this docile material.

85 REINFORCED CONCRETE

Bridge over the Ourthe, Liege, Belgium. Built during four winter months for the
Liege Exposition of 1905. Francois Hennebique, engineer. 180-foot span.

Hennebique (1842-1921), a French engineer-contractor celebrated for his early
development of reinforced concrete construction, was one of the first to realize that the
new material lent itself to a smooth flow of structure and surface.

In its lean elegance this bridge has had few rivals. The flattened arch becomes
amazingly thin at the crown, yet this photograph taken on the official proving day
shows that it was capable of supporting three steam-rollers.

REINFORCED CONCRETE 86

Tovonasa Bridge over the Rhine. Canton Grisons, Switzerland. 1905; destroyed by
landslide in 1927. Robert Maillort, engineer. Three-hinged arch of 167-foot span.

Following his master, Hennebique, in the quest for integrated structure, Maillart
fused arches and road slab to form a structural unit, proudly revealed. Further
explanation of the construction will be found on page 106.

This was Maillart's first masterpiece.

87 REINFORCED CONCRETE

REINFORCED CONCRETE ARCH

Detroit-Rocky River Bridge, neor Cleveland, Ohio. 1911. 280-foot span, g
Arch ribs are plain concrete, but the rest of the bridge is reinforced
Like the other first large American arches in the new material, this
was closely patterned after a famous stone bridge of 1903:
the Pont Adolphe at Luxembourg, a twin-ribbed open-spandrel arch
of record 280-foot span. The more appropriate models of Hennebique
and Maillart (pages 86 and 87) were disregarded then as now.

Bixby Creek Bridge on the Carmel-San Simeon Highway, California.
1933. By the California Division of Highways. 320-foot span.

The great orch ribs are dwarfed to insignificance by colossal
abutment piers, a misplaced emphasis that distorts the balance of the JKJ^S
bridge and destroys its continuity of line. Compare with the more
recent bridge by the same office that is shown on the facing page.

Thickened piers over arch abutments are generally rationalized as
wind bracing, but since they hove been proved dispensable their
continued popularity in this country seems attributable to nostalgia for
the monumental forms of ancient stone construction.

It is in arches that reinforced concrete achieves its boldest spans. Occasionally the
roadway is suspended from the arch. More often it runs above, as in the bridges shown
on the next few pages. Since plain concrete masonry, like stone or brick, is strong in
compression, little or no reinforcement is required for a massive fixed-end arch, but
only skillfully embedded steel makes possible limber two and three-hinged arches,
slender spandrel columns or cross-walls, and thin flat decks.

Typical of today's best standard practice are the graceful arches shown on the
following pages. They are so undeniably handsome that it is perhaps ungrateful to
complain of their somewhat indifferent relationship to their material. Assembly of
apparently separate pieces— supporting ribs, intermediate posts and supported, seem-
ingly inert deck— conveys little feeling of the unity and continuity of structure and of
shape that is implicit in reinforced concrete, and represents an advanced stage of the
development toward lighter and more economical masonry construction that started
with the medieval Ceret arch (page 24) rather than design freshly conceived in the
specific terms of a totally new material. A typical offender in this special sense is the
otherwise admirable Russian Gulch Bridge (opposite). Compare it with the work of the
independent master, Maillart (pages 87 and 102-1 13), where steel and masonry, arch
and superstructure, are so completely fused into a single working shape that execution
in any material other than reinforced concrete is unthinkable.

REINFORCED CONCRETE ARCH 88

Russian Gulch Bridge on the Mendocino Coast Road south ot Fort Bragg,
California. 1940. By the California Division of Highways:
F. W. Pgnhorst, bridge engineer. 240-foot span.

The valley is spanned in one graceful gesture, for the posts that
support the roadway march without break over banks and arch, waning
in size as they approach the crown and waxing as they take the
downward path. The complete separation of arch from roadway is very
pleasing in this high-reaching elliptical arch, whereas flatter arches
look best when they are joined with the roadway at the crown.

We are so accustomed to considering good spare concretework as a
prohibitively expensive luxury in the United States that it is gratifying
to hear that this beautiful bridge was judged the most economical
solution to the problem.

89 REINFORCED CONCRETE ARCH

Sandobron (Sando Bridge) over the Angermon River, Sweden. Built
1937-42. By the Skdnska Cement Company: S. Haggbom,
chief enqineer. 866-foot span.

This is the longest reinforced concrete arch in the world.

Pairs of round columns support the long approach viaducts (see also

page 116) and lift the roadway over the mammoth single-ribbed arch.

Traneberg Bridge, Stockholm. See opposite page.

REINFORCED CONCRETE ARCH 90

Traneberg Bridge, Stockholm. 1934. For the Stockholm Harbor Board:
Ernst Nilsson and S. Kasarnowsky, engineers; Eugene Freyssinet,
consultant; Paul Hedquist and D. Dahl, architects. 585-foot span.

The transverse slab-walls that carry the deck obstruct the diagonal
view, giving an illusion of mass, yet they seem more appropriate to
reinforced concrete than the isolated posts of the Sando Bridge on the
opposite page, and remove the otherwise startling resemblance to

steel construction. These cross-walls carry the roadway smoothly over the
twin-ribbed arch and beyond, interrupted only by the fusion of deck
and arch at the relatively low crown. There is no special monumental
treatment at the abutments.

Note the crisp molding about the upper edge of the arch ribs, also
the cleanly cantilevered sidewalk with its light steel railings.

91 REINFORCED CONCRETE ARCH

Kungsbron, Stockholm.

A close view of one of the flat-arched
twin-ribbed spans of the unusual double bridge
that is shown on the opposite page.

Royal Tweed Bridge, Berwick, England. 1928.
L. C. Mouchel and Partners, engineers.

The rhythmic continuity of the long flat
arches and the spandrel posts is unbroken by
any special accent over the abutment piers. Note
that the span of the arches increases as the
roadway mounts from the low bank on the right
to the higher bank on the left.

The juncture of arch crown and deck, never
an easy problem, seems somewhat tentative as
compared with the Swedish bridges, but the
treatment of spandrel walls and posts as a
smooth continuous plane, recessed behind deck
and arches, has considerable merit.

REINFORCED CONCRETE ARCH 92

Kungsbron (King's Bridge), Stockholm, c. 1940. For the Stockholm Harbor Board:
A. Wickert & S. Kasarnowsky, engineers.

Since o broad-decked short-span bridge is bound to look stubby and awkward, this
small urban bridge was molded in two completely separate sections, each carrying
one-way traffic. The longitudinal split also allows the bridge to fit easily into its
man-scaled surroundings.

93 REINFORCED CONCRETE ARCH

Gueuroz Bridge over the Trient Glacier, Canton Valais, Switzerland.
1933. A. Sarrasin, engineer. 323-foot span.

Reinforced concrete has been cast into a working shape of
extraordinary visual power, closer in spirit to the work of the older Swiss
engineer, Maillartfpages 102-113), than to the other arches shown
in this section.

Parapets are usually inert, extraneous elements, but here they serve

ds beams to support the approaches and to stiffen the slender arcn
ribs. This interdependence of structural members is expressed in the
smooth flat plane formed by the parapet, posts and arch.

The refinement of line and proportion is all-pervasive. Note the
vigorous shape of the voids formed by the rounded juncture of posts and
parapet, and the fine relationship of the spandrel posts to the sturdier,
more widely spaced verticals that carry the approach spans.

REINFORCED CONCRETE ARCH 94

Gueuroz Bridge, Switzerland.

A worm's-eye view of the distinguished bridge that is illustrated on
the opposite page.

The spacing of the braces that tie together the two arch ribs has
been handled with exceptional neatness.

Proposed bridge over the Rhone at St. Maurice, Switzerland, c. 1945.
A. Sarrasin, engineer. 328-foot span.

The amazing thinness of the deck is due to its unusual construction
as an active self-supporting slab of reinforced concrete, growing out of
the rectangular mushroom-headed columns that carry it over arch
ribs and river banks. There are no beams.

This type of mushroom-slab construction, with its smooth flow of line
from column into slab, was invented by Maillart around 1908 (see
page 102) and used by him in buildings of many kinds— never, however,
in his bridges. Maillart preferred to reveal his thin slabs as such only
in the substructure, and thus kept the apparent weight of his bridges high,
with gratifying results. His stiffened slab-arches (pages 102-104),
quite the reverse of this project in principle, have nothing of its
droopiness. Their emphasis is upon the firm continuous line of the
roadway itself, not upon the means of its support.

Another notable feature of this design is the use of paired
columns for wind-bracing at the abutments instead of the customary and
ungainly thickened piers.

The United States has a bridge of similar construction— the
Fort Snelling-Mendota Bridge built over the Minnesota River in 1926,
designed by C. A. P. Turner and Walter H. Wheeler. But it is far less
graceful than this Swiss project, partly because it uses Turner's system of
mushroom-slab construction (patented in 1905) in which columns are
separated from the slab by rectangular plinths or capitals.

95 REINFORCED CONCRETE ARCH

La Roche-Guyon Bridge over the Seine, France. 1934; destroyed in
World War II. Etablissements Boussiron, engineers. 528-foot span, a
record for overhead reinforced concrete arches.

Even in a country with a long tradition of fine reinforced concrete
construction this arch was of outstanding merit.

Arch ribs are of smoothly varied section: at their spring they are flat
hollow rectangles; at road level they are solid and square; at the crown
they again become hollow rectangles, but now emphatically vertical.
The crescent shape, unusual in hingeless arches, is particularly effective
in the slow-rising curve used here, and the light lattice-bracing

avoids any feeling of top-heaviness.

Note that the hangers (pure tension members) are of reinforced
concrete and designed to match the posts (compression members) that
support the approaches. This note of formalism should probably be
condemned, yet, as in the Chicago overpass pictured on page 53,
it contributes a great deal to the unity and harmony of the design. Light
steel hangers would have been more reasonable, but in that case
the visual center of gravity would have slipped below the roadway
and upset the balance of values.

REINFORCED CONCRETE ARCH 96

Bridge over the Kalix alv, northern Sweden. 1933. Skdnska
Cement Company, engineers.

The overhead arch as an impressive demonstration of the
power of masonry.

The emotional impact of the overhead arch is unique. The more sharply defined and
proudly isolated, the more monumental its effect. It is not so much an assertion of
reinforced concrete, for steel plays a very minor role in its compressed curve, as it is
the ultimate demonstration of the brute power of plain masonry. The United States offers
nothing comparable in excellence with the European examples illustrated here, for the
American designer, instead of playing up the lightness of hangers and roadway and
railings as legitimate contrast to the massive supporting arches, tends to extend the
weightiness of his arch ribs to each smallest detail and thereby kills the spirit of his
structure.

Bridges with overhead structural members— suspension bridges excepted— are a
difficult problem in any material, for there is always the danger that vital cross-braces
will seem either distractingly complicated or oppressively bulky. Compare with the steel
overhead arches and trusses shown on pages 50 to 53.

97 REINFORCED CONCRETE ARCH

Bridge over the Seine at St. Pierre du Vauvray, France. 1922; destroyed in World War II.
Eugene Freyssinet, engineer. 430-foot span, then a record for reinforced concrete.

To pass over this famous bridge must have been a startling experience, for the
roadway was scarcely wider than the sum of the two arch widths. One would first
have felt squeezed between the giant ribs, then liberated as they soared up through
space, free of cross-bracing other than the heavy frames at either portal.

Suspended from the hollow cellular arch ribs were the wire hangers, thinly coated
with cement mortar, that supported the light trusses that in turn carried the
roadway. The filigree of the trusses was recalled in the design of the railings.

Sorangsbron (Sorangs Bridge), Halsingland, Sweden. 1930. Skanska
Cement Company, engineers. Three-hinged arch of 177-foot span.
Inclined hangers, a Scandinavian invention, reduce the stresses
in the arch ribs and permit their extreme slenderness.

The three-hinged arch was cast as two entirely separate pieces.
The exceptional neatness and lightness of arches, hangers and deck
make the awkward design of the abutments seem all the more
unfortunate by contrast.

REINFORCED CONCRETE ARCH 98

Proposal for a thousand-meter (3281 -foot) arch of reinforced concrete.
Eugene Freyssinel, engineer.

The substantial reputation of its designer lends authority to this daring project.
In 1928 Freyssinet proposed such a bridge for the Hudson River.

Divided, splayed haunches contribute to the grace of the tapered curves, and the
level at which the roadway and arches intersect seems very well chosen.

,'f

99 REINFORCED CONCRETE ARCH

Prestressed concrete bridge of Luzancy. See opposite
page.

"The hollow rib that is being hoisted into place in
this picture is made up of many short precast concrete
sections, compressed by taut steel wires to take up
dead load. No scaffolding is necessary. When the rib
is in position longitudinal wires will be threaded through
from one abutment to the other and tightened to pull
the sections into compression and form an arch.
The total prestressing will allow for moving loads as well
as for the weight of the bridge itself.

Supports for a third rib are evident at extreme left
and extreme right. The thrust is transmitted to the
massive abutments through the split arch ends. Because
of this articulation at the abutments the bridge is
called a ponf a bequilles, or "crutch bridge."

As sometimes happens in technically advanced
reinforced concrete construction, the various elements
of the bridge are interwoven in such complex fashion
that its exact structural nature is somewhat controversial.
It may be termed an arch, but from another point
of view it might better be described as a rigid frame
of very unusual type.

Prestressed concrete bridge at luzancy.

An advanced stage of construction. The upper
and lower surfaces of the three hollow ribs have
been joined with precast concrete slabs and
transverse wires.

The completed bridge is illustrated on the
opposite page.

REINFORCED CONCRETE ARCH 100

Prestressed concrete bridqe at luzancy, Seine and
Marne, France. 1946. Eugene Freyssinet, engineer.

The leadership of France in reinforced concrete construction has been due pre-
eminently to Hennebique (page 86), to Auguste Perret, the architect, and to Eugene
Freyssinet, celebrated for his vaulted hangars of 1924 at Orly, his great arched bridges,
one of which is shown on page 99, and now for his work with prestressed concrefe.

The prestressing of concrete consists in artificially creating stresses approximately
equal and opposite to those that are produced by dead weight and live load in the
completed, functioning structure. In the Freyssinet system this is effected through the
pressure exerted by a series of parallel steel wires of high tensile strength that are
stretched to the limit of their elasticity and embedded in the concrete, thus creating
permanent compression. Such construction uses at least 70% less steel than ordinary
reinforced concrete, an economy of financial importance; and it uses 30 to 40% less
concrete, an economy of esthetic importance, for it implies unprecedented slenderness.

Freyssinet advocates the use of prestressed concrete in prefabricated sections.
Entire factory-made beams are feasible for small and medium spans. Long beams or
arch ribs, however, are assembled on the site from precast concrete units of easily
handled size. These are placed end to end and joined by wire threaded through holes
provided for the purpose. The joints between units are then filled with mortar and the
cables stretched taut. When the beams or arches, whether one-piece or composite, are
all in place, they are threaded together transversely with more stretched wires.

The system found its first major proving ground in wartime Tunisia, where June, 1 943,
found the reinstated French with 300 bridges that needed immediate repair or replace-
ment, and with very little steel and almost no wood for forms. Prestressed concrete in
prefabricated units was the logical answer and proved successful far beyond its
emergency value. The Tunisian bridges were not particularly attractive, but the visual
possibilities of the construction are suggested by the handsome new bridge at Luzancy
that is pictured here.

101 REINFORCED CONCRETE ARCH

Vol Tschiel Bridge, Canton Grisons, Switzer-
land. 1925-26. Robert Maillart, engineer.
Stiffened slab-arch of 142-foot span.

Here for the first time Maillart used his
extremely thin barrel vault, stiffened by the
deep rigid girder formed by parapets and
road-slab.

The semicircular openings along the
parapet and the incongruous masonry
abutments were stipulated by the Grisons
officials.

Robert Maillart (1872-1940) used reinforced concrete to enter wholly new realms
of structure and shape. This Swiss engineer was so far in advance of his time that the
full meaning and impact of his work may not be felt for years to come. Whereas most
engineers tend to be enslaved by their formulas, Maillart used his formidable scientific
and technical knowledge as the tool of his intuition, creating structure that transcended
accepted patterns and limitations to reveal the laws of nature in terms of new and sur-
passing beauty. It is not surprising that the forms at which he arrived with notable
independence should be kin to those evolved by other great modern artists working in
other and very different media.*

Early experience under Francois Hennebique, French engineer famous for his
pioneer work with reinforced concrete (page 86), must have sharpened Maillart's own
awareness of the plastic character of the new material and its irrelevance to struc-
tural shapes traditional in stone or metal. Out of his profound understanding of rein-
forced concrete he gradually developed new and magnificently appropriate types of
construction in which each small part became an actively participating member of an
organic whole.

Although Maillart designed many buildings, his greatest achievement was his
bridges. Most of them are small and hidden in remote Alpine valleys, for his work was
too unconventional to receive much support; but his spans were relatively so inexpensive
—because of their efficient use of material— that officials could not afford to ignore him
entirely. These bridges are in the main of two distinct types: the stiffened slab-arch,
illustrated here, and the three-hinged arch with integrated road-slab, shown on pages
106 to 1 1 1. His great invention of the mushroom slab** he did not himself apply to
bridges, but a proposed application by a younger engineer is illustrated on page 95.

*A stimulating study of these relationships, outside the scope of this book, may be found in Space, Time and Architec-
ture, by Dr. Sigfrled Giedion.

"The most complete presentation of Maillart's work will be found in a monograph by Max Bill that has very recently
been published in Switzerland.

REINFORCED CONCRETE ARCH 102

MJ

Maillart's stiffened slab-arch is something like a reversed suspension bridge in its
structural action, for the flexible vault, of eggshell thinness, takes only direct thrust,
while road-slab and parapets together form a rigid U-shaped girder that resists local
bending under concentrated moving loads. Spandrel supports, too, are thin continuous
slabs, active in all three dimensions. Characteristic of Maillart's use of reinforced con-
crete is this emphatic insistence upon the slab as the basic element of construction, far
more appropriate to the material than the usual steel-inspired network of isolated posts
and isolated beams.

The miraculous lightness of these bridges must be attributed to the extraordinary
efficiency of their revolutionary construction, their equally miraculous elegance, to the
consummate artistry of their designer.

Footbridge near Wulflingen, Canton Zurich. Switzerland. 1933.
Robert Maillort and W. Pfeifter, engineers. Stiffened slab-arch of
1 24-foot span.

The subtle reverse curve of the footway makes this one of the most

graceful of Maillart's bridges. Arch and deck slabs fuse into one as
they approach mid-stream, yet the total thickness at the crown is only
4Vj inches.

103 REINFORCED CONCRETE ARCH

Schwandboch Bridge. See opposite page.

The inner edge of the slab-arch follows
the curve of the highway.

Schwandboch Bridge.

The outer edge of the arch, straight ir
| plan, serves as base for the sloping sides
V'A of the cross-walls that support and brace
the curved deck girder

i

REINFORCED CONCRETE ARCH 104

Schwondbach Bridge, near Schwarzenberg, Canton Berne, Switzerland. 1933.
Robert Maillart, engineer. Stiffened slab-arch of 111-foot span.

A curiosity among bridges, this curved span is a dramatic example of Maillart's
extraordinary feats of engineering. As demonstrated by the photographs on the
preceding page, the inner edge of the vault follows the elliptical curve of the
deck, but the outer edge is straight, with vertical cross-walls brought up on a
diagonal to buttress the bridge against centrifugal action.

The thinness of the arch (7.9 inches) and the cross-walls (6.3 inches) looks
precarious to anyone unfamiliar with the fantastic strength of reinforced concrete in
favorable construction. A special odvantage of the extremely light vault of these
bridges is the minimum of expensive scaffolding that is required for support
during construction

05 REINFORCED CONCRETE ARCH

• v.

Kj- ^j.

r3^-~lg^'v^^J^i?!2T>mjM

H|% ^^ * \T>

ri •'"'/-■/ ' *_-JKm

i^jH

m£M^WT

►*.:hwhKiKiSF-.'*: >• ' . - ->"-"7t7a"7

t*-^I§v9e£r '^ *' *3' J/ /

v^SI

^WlN^^W

p'^gH KS*, *S^±>r>ify%-

mMWrnkWrn

~>'^%ssl

V vkVI .1 *v!5»S^S'tt«E:M? -

RtSBIV: , ,%al*H-4KsSr ?:>

Kh'JBbU > *■ * -flf-lPH

i»si|r ^■^RBfc'Mfc*

yra«*w "¥lH«nad

fck- ^h^hH

•BSBK*1 fHmB Iko

rw' i$WKs»' ^liflHk ^f^k^l

d(a ^fci?£— ' fjg^lB

9l,. whU B^vKI

5kJ

R, ;i*2SM Ek- sfl

K ^H RSI

H^ *v9

IP^sPPi

K&^ *v**k---^HB

Bl ■ ■ fO^*ai

The scaffolding of the Salgino Bridge. See opposite page.

The daring lightness and notable elegance of the arch were presaged
in its scaffolding, designed and executed by the Coray family of Chur, long
famous in this highly specialized field.

Maillart's other great bridge type was the three-hinged arch with integrated road
slab, suited through its elasticity to greater spans and to less stable foundation condi-
tions than the stiffened slab-arch illustrated on the preceding pages.

Joints at mid-span and at either abutment divide a three-hinged arch into two
symmetrical halves, each of which should be thickest at its center if it is to make most
efficient provision for moving loads. With its bulging ribs under a separate, passive
roadway, the usual arch of this type is singularly ungainly. Even while he reduced its
weight and heightened its effectiveness, Maillart transformed the three-hinged arch
into a thing of beauty. As early as 1905, in his Tavanasa Bridge (page 87), he used
a reinforced concrete road slab as an active structural member, fusing it with his open
U-shaped arch ribs at the critical quarter points to form a strong, closed box-shaped
girder that tapered to the hinge at the crown.

He developed innumerable variations upon this theme. The solid masonry abutment
piers of the Tavanasa arch were replaced by thin cross-walls of reinforced concrete,
similar to those he used as supports between arch ribs and deck girder. In his latest
bridges these transverse slab-walls, and the arch ribs too, assumed vigorously curved
and angled outlines as Maillart shaped them to extract the utmost strength and meaning
from his material.

The significance of these bridges goes deeper than the lithe elegance of their
appearance or the technical virtuosity of their structure, for in them, by grace of their
creator, reinforced concrete is quickened to life and given a voice unmistakably its own.

REINFORCED CONCRETE ARCH 106

Solgma Bridge, near Schiers, Canton Grisons, Switzerland. 1930.
Robert Maillart, engineer. Three-hinged arch of 269-foot span.

A classic version, of its structural type, the lean and flattened arch
bridges the chasm in one smooth leap.

107 REINFORCED CONCRETE ARCH

Bridge over the Aare at Innertkirchen,
Canton Berne, Switzerland. 1934.
Robert Maillart, engineer. Three-hinged arch of
96-foot span.

This small bridge is of different construction
and rather gentler demeanor than Maillart's
customary three-hinged arches. His usual
U-shape is reversed here, for spandrel walls
and road slab together form a girder that is
open beneath.

Bridge over the Thur, near Felsegg, Canton
St. Gall, Switzerland. 1933. Robert Maillart,
engineer. Two parallel three-hinged arches of
226-foot span.

No bridge of Maillart's is more assertive
of strength than this light span over the Thur.

Characteristic of his later work are the
powerful, concisely defined, highly differen-
tiated shapes — the pointed arch, the straight-
drawn outer edges of the arch ribs, the splayed
slab-supports of the approaches. Every part is
alive and at work.

The X-shaped abutment-joints of reinforced
concrete, more economical than conventional
steel hinges, contribute a great deal to the
sense of unity ond continuity of structure.

REINFORCED CONCRETE ARCH 108

Bridge over the Thur. See also opposite page.

At once bold and delicate, the lithe arch looks very much at ease in
the friendly, man-scaled landscape.

109 REINFORCED CONCRETE ARCH

Bridge over the Arve, near Geneva,
Switzerland. 1937. Robert Maillart, engineer.
Three parallel three-hinged arches of
194-foot span.

In paring the substance of his bridges to an
irreducible minimum, Maillart evolved
extraordinary new forms. Like the abutment
hinges that they so much resemble, these
X-shaped road supports combine elasticity and
strength with economical use of material.

The light railing, assembled of rolled steel
sections, is a model of propriety.

Lachen Bridge, Altendorf, Canton Zurich,
Switzerland. 1940. Robert Maillart, engineer.
A skew bridge with two separate off-set
three-hinged arches.

Arches spring from different levels to carry
a highway over railroad tracks at a sharp
angle, and the static symmetry that we take for
granted in a bridge span is replaced by a
dynamic interplay of shapes.

Bridge over the Simme, Garstatt, Canton Berne,
Switzerland. 1939. Robert Maillart, engineer.
Three-hinged arch.

The smooth curves of the Salgino arch
(page 107) hardened nine years later into these
taut diagonals— strong ond decisive, infinitely
expressive.

The gabled roof in the background belongs
to an old wooden bridge.

REINFORCED CONCRETE ARCH 110

Bridge over the Arve. See also opposite page.

The very slightly curved arch ribs meet in a point at the crown and the bridge
becomes wholly expressive of its tri-jointed construction.

Ill REINFORCED CONCRETE ARCH

REINFORCED CONCRETE: BEAM AND RIGID FRAME

Gijndlischwand Bridge, Canton Berne, Switzerland. 1937. Robert Maillart, engineer.
A skewed continuous beam with main span of 125 feet.

Maillart's bridges were not invariably arches, but they were always of reinforced
concrete; ond whatever their structural principle they were imbued with their
designer's acute awareness of the unique nature of his chosen material.

REINFORCED CONCRETE: BEAM AND RIGID FRAME 112

Chatelard Aqueduct, Canton Valais, Switzerland. 1925. Robert Maillart, engineer.
100-foot span.

The structure is hybrid, for the arch springs conventionally from its abutments, then
merges with the box girder that carries the water.

This fusion of two seemingly incompatible forms is curiously successful.

113 REINFORCED CONCRETE: BEAM AND RIGID FRAME

s±J ukUbk

Proposal for a long-span highway bridge of reinforced concrete. 1948.
Paolo Soleri, architect. Continuous beam.

The undulated slab flies over the river like some strange sleek bird. Its winged
flanges are convex at the piers, then soar up and over in a reverse curve to
embrace the roadway at midspan. There are no separate elements-only the
attenuated multi-curved slab, one with the piers from which it springs.

Essentially the bridge is a tube-carved away where superfluous and turned
inside out at the piers.

Maillart showed how a reinforced concrete bridge might become one thing, how
it might grow out of the fluid, continuous character of its material. It was in this

spirit that he developed his principle of the slab Perhaps it had to be an architect,

committed to the creation of space by the very nature of his art, who would take
the next step and free the slab to come alive in three full dimensions.

REINFORCED CONCRETE: BEAM AND RIGID FRAME

1 14

Proposal for a long-span highway bridge. See also opposite page.

Sketched above is a lineal analysis.

Reproduced below are elevation, plan, longitudinal section and, at bottom
right, the transverse sections at mid-span, quarter-span and mid-pier.

■—■ '-'-■"'" --p^WM^Mpp

^

15 REINFORCED CONCRETE: BEAM AND RIGID FRAME

Bull Run Creek Bridge for the Norris Freeway,
near Knoxville, Tennessee. 1934. By the
Tennessee Valley Authority. Continuous beam
with main spon of 50 feet.

Through the flat planes of piers and beams
the bridge becomes a geometric abstraction of
its structural idea.

Approach to the Sando Bridge, Sweden.
(See page 90.)

The curving viaduct is so high, so light,
so cleanly drawn that its unobtrusive presence
actually enhances the quality of the natural
landscape.

Bridge over Henderson Say, Pierce County,
Washington. 1937. By Pierce County:
F. A. Easterday, engineer. Cantilever beam with
main span of 190 feet.

The road slab forms the top of a hollow box
girder, and the girder in turn is monolithic
with its supports. Note the "suspended span"
inserted at center.

The designer has stated his principle of
construction in unusually agreeable terms.

Dry Creek Bridge, Wabunsee County, Kansas.
1941. By the Kansas State Highway Department:
E. S. Elcock, designer, G. W. Lamb, bridge
engineer. Rigid frame with spans of 50,
70 and 50 feet.

The substructure, with its tapered, divided
piers, is shaped with exquisite skill, but the
design as a whole suffers from the overcomplica-
tion of coping and parapet.

Side spans run straight to the abutments,
with no semblance of arch construction.

REINFORCED CONCRETE: BEAM AND RIGID FRAME

1 16

Overpass at Oelde, Germany, c. 1938. For the Autobahn:

Karl Schaechterle and Fritz Leonhardt, chief engineers; Paul Bonatz,

architect. 108-foot span.

This simple beam is comparable to the Nazi plate girder bridges
(pages 65 and 66) in its sober refinement and in such specific details as
the shallow, decisively projected sidewalk slab, continued over the
retaining walls as a coping, and the light railing without terminal
accents.

The ends of the beam are not concealed. They rest in full sight upon

their supporting piers, contributing to the notable clarity of statement.

The precast beam looks rather heavy, especially considering the
fact that it was prestressed according to a variation of the Freyssinet
system described on page 101; but then the Germans have never had
the light touch with reinforced concrete that has been so characteristic
of French work.

It is with some justice that flat-spanned overpasses of this type have
been criticized as a psychological obstruction to fast traffic. Arched
openings usually seem higher and safer to a speeding motorist.

Straight beams are reasonable and economical in reinforced concrete, though
inexpressive of its special character. Design problems are much the same as in similar
construction of steel, but elegance is more difficult to attain, for reinforced concrete is
relatively bulky and lacks the vertical stiffeners and the sharply profiled edges that
give delicacy and scale to a steel plate girder. But brutality can be avoided through
precise statement of the structural principle, and through the welcoming of every oppor-
tunity to introduce lightness and fineness as contrast to the dominant mass.

Rarely at complete ease in rectilinear forms, reinforced concrete comes into its own
in continuous beams with curved under-edges (see the discussion of such construction
on page 10) and in rigid frames, where the fusion of verticals and horizontal is particu-
larly well suited to the fluid quality of the material. In either event the construction
assumes its own logical form, which it can do very handsomely, without trying to dupli-
cate the appearance of arches.

117 REINFORCED CONCRETE: BEAM AND RIGID FRAME

Gardiol Bridqe above Montreux, Switzerland.
1944, E. Gardiol, engineer. Continuous
beam with 48-foot spans.

Like many mountain bridges, this narrow
railway viaduct is curved in plan. Its con-
tinuous beam construction is of an unusual
type, for the beam is not set on rollers,
but cast as one with the slim splayed piers,
some of which are 98 feet tall. These flexible
supports provide the elasticity needed to
allow the beam to move in response
to temperature changes.

Waterloo Bridge, London. See also opposite
page.

REINFORCED CONCRETE: BEAM AND RIGID FRAME 118

Waterloo Bridge over the Thames, London. Built 1939-45 to replace
John Rennie's famous Waterloo Bridge of 1817. Rendel, Palmer & Tritton,
engineers, in association with Sir Peirson Frank; Sir Giles Gilbert Scott,
architect. Continuous beam with five 240-foot spans.

Long leaping curves are executed with such easy grace that the great
new bridge, far from disfiguring the ancient face of London, brings
it new life, new and exciting perspectives. The dome of St. Paul's
dominates the skyline at the right. Here is ample proof that distinguished
twentieth-century architecture can take its place proudly in any
setting. Compare with the manner in which the designers of our

Arlington Bridge in Washington (page 85) solved their somewhat
similar problem.

Neither the reeded coping nor the angular motif at the junction of
the beam and piers is completely convincing, but the latter may be
partially accounted for by the unusual construction: the continuous beams
are not set on rollers to allow for movement, but fused with flexible
bearing walls that are set within rigid shell-like piers.

The bridge is faced with slabs of Portland stone, laid in vertical
courses to avoid any resemblance to solid masonry.

19 REINFORCED CONCRETE: BEAM AND RIGID FRAME

Proposed highway bridge over the Wisconsin River near Spring Green, Wisconsin.

1947. Frank Lloyd Wright, architect. Cantilever beam adaptable to spans up to
200 feet.

The architect calls it a "butterfly" bridge because its outstretched wings
concentrate the load upon a deep central girder. The elegance of the design is
best evident in the cross section at mid-span that is illustrated above: at this point
the substance of the bridge is reduced to the shallow V-shaped structure shown in
solid black, while the shaded portions indicate the increasing depth of the span
as it curves back to the inset piers from which it springs. The heavy longitudinal
girder at the center projects above the deck to separate the traffic lanes.

The outer shell is no inert surface, but a "stressed skin" that works as one with
the light stiffening ribs. Structure becomes continuous, flowering out of the plastic

nature of the material, consistent with Frank Lloyd Wright's conception of

architecture as organic.

The cantilever principle is used lengthwise as well as crosswise. The drawings

on the facing page show that the bridge is conceived as a series of standardized

self-supporting units, each cantilevered out from its central pier to meet the

arms of adjacent units at mid-span.

Most engineers' bridges simply cut through space. Their interest is in flat elevation

rather than in the depth plane. This architect's bridge is quite a different matter,

for it gives space shape and meaning.

REINFORCED CONCRETE: BEAM AND RIGID FRAME 120

Proposed bridge over the Wisconsin River. See opposite page.

The drawings show the plan and elevation of two
identical, adjacent units. In the plan one of the units is
partially cut away to reveal the pier and stiffening ribs
beneath the deck.

,

1
\r'"\ ',""■ ' r"~ ;r'--ir" " r~?

\ 11 :;. ■ / \ li '■?■

Uj !j !| \ / !;' •

t-"

"' ; 1 l \ / ''

. 'hr-ir ;r'

\ i;

h, fSar'- • ilk

121 REINFORCED CONCRETE: BEAM AND RIGID FRAME

Proposal for a Butterfly-wing Bridge over San Francisco Bay.

This plan shows the division of traffic into two separate highways as
the bridge makes its great triple jump over the main channel of the Bay.

The two arcs are joined at the center, 175 feet above water, by
gardens and parking space disposed on a platform of reinforced
concrete.

Proposal for a Butterfly-wing Bridge over San Francisco Bay.

Great shell-like cantilever arms span the three broad openings
making a record-breaking thousand-foot leap at the center.

REINFORCED CONCRETE: BEAM AND RIGID FRAME 122

Proposal for a Butterfly-wing Bridge over San Francisco Bay from
San Francisco to Alameda, California. 1949. Frank Lloyd Wright,
architect; J. J. Polivka, engineer. Cantilever beam with typical span of
156 feet; main spans of 500 and 1,000 feet.

The butterfly-wing principle that Frank Lloyd Wright first conceived
for the modest requirements of the Wisconsin River (see the two
preceding pages) comes into full flower in this proposal for a southern
crossing over San Francisco Bay. The site is the one favored by
most authorities as the best means of easing congestion on the existing
Bay Bridge.

The body of the bridge is composed of a single repeated cantilever
unit similar to that of the Wisconsin project. The roadway is balanced
upon a central longitudinal girder that grows out of the tap-root piles.
Its spread is reinforced from beneath by thin shells of concrete-sprayed
steel mesh that curve up and out from the lower edge of the deep
spinal girder.

The soaring double arc of the roadway as it splits and swells
outward and upward over the main channel is a brilliant variation and
expansion of the established, oft-repeated theme. It is not a true

arch, nor does it simulate one. Instead, it is formed by the out-reaching
of cantilever arms from the two great U-shaped piers to meet at
mid-span, 175 feet above the water. Here the two lightly flying, out-curved
halves of the roadway are joined and braced by a landscaped park,
fabulous hanging gardens for the delight of citizen and sightseer.

The bridge would be extremely economical in construction and
maintenance. Erection of the spans as stiffly reinforced arms obviates
much of the costly erection work usual in reinforced bridges, for the stiff
reinforcement itself serves as scaffold and centering. The small
standardized spans are well adapted to prefabrication, and what little
formwork is needed might be used again and again. There would
be little maintenance, for there would be no exposed steel to paint and
replace. A further advantage claimed by the designers is relative
earthquake safety— a claim worth attention when made by the architect
of the Imperial Hotel in Tokyo.

The longest span now achieved in reinforced concrete is the 866 feet
of the Sando arch in Sweden (page 90). If the people of the Bay
region have the foresight to translate vision into reality, they will have
a bridge second to none in the world in beauty and in boldness.

g — ~. -., , >?^S _~ ':— '"r-t-

123 REINFORCED CONCRETE: BEAM AND RIGID FRAME

GLOSSARY

abutment.- An end pier of a bridge, particularly of an arched bridge,
arch: See Structural Types, page 1 1 ; also for fixed arch, two-hinged arch and three-
hinged arch.

segmental arch. An arch curve that forms part of a circle.

elliptical arch: A curve determined by two foci.
bascule bridge-. A drawbridge working on a horizontal pivot,
beam. See Structural Types, page 10; also for continuous beam,
caisson. A box or chamber used for construction under water.
cantilever: See Structural Types, page 10.
coping.- The capping or covering of a wall.

corbel- A projection from the face of a wall, supporting a weight,
cornice: Moldings run along the top of a wall,
crown.- The apex or summit of an arch,
cutwater.- An angular or curved structure projecting from a pier that cleaves the water

and so lessens its pressure.
girder: A supporting horizontal beam.

plate girder: A solid-walled metal girder.
laminate-. To build up of separate laminae or layers.
lintel: A horizontal beam supporting an opening.

modi/lion: An ornamental block or bracket under a projecting cornice,
monolithic.- One-piece structure: material or materials so brought together as to become

an indissoluble structural unit.
parapet: A low wall or protecting railing.
pier-. One of the vertical supports of a bridge.
rigid frame: See Structural Types, page 10.
spandrel: The walls between supporting vault and bridge-deck.

truss-. Separate members (such as beams, bars or rods) assembled to form a rigid
framework.

vault.- An arched structure.

125

SOURCES OF ILLUSTRATIONS

STONE

p. 12, E. Jervoise, the Notional Buildings Record,
London; p. 13 (above) Emilia Bologna, courtesy the
Harvard Architectural Library, Cambridge, Mass.;
(below) Kunst im Deutschen Reich, Vol. 3, No. 8,
Aug., 1939; p. 14, Photo-Molina, Black Star, N.Y.j
p. 15, Folger Shakespeare Library, Washington,
D.C.; p. 16, E. Jervoise, the National Buildings Rec-
ord, London; p. 17, Philip D. Gendreau, N.Y.; p. 18,
Alinari, Florence; p. 19, Gauthey, Emiland Marie:
Oeuvres Traite de la Construction des Ponts, Novier,
ed., Paris, Didot, 1809, Vol. 1, fig. 3; pp.20 and 21
(above) Palladio, Andrea: The Architecture of A.
Palladia, London, Ward, 1742; p. 21 (below) National
Buildings Record, London; p. 22, Philip D. Gendreau,

N.Y.; p. 23, E. Meerkamper, courtesy Swiss Federal
Railroads, N.Y.; p. 24 (above) Gauthey, Emiland
Marie: Oeuvres- Traite de la Construction des Ponts,
Navier, ed., Paris, Didot, 1809, Vol. I, pi. 64;
(below) Duplication Service, Library of Congress,
Washington, D.C.; p. 25 (above) Theory, Practise and
Architecture of Bridges, John Weale, ed., London,
The Architectural Library, 1843, Vol. IV, pl.58; (be-
low) Philip D. Gendreou, N.Y.; p.26, courtesy Dr.
Ssu-ch'eng Liang; p. 27, E. Jervoise, the National
Buildings Record, London; p. 28, Berenice Abbott,
N.Y.; p. 29, Kunst im Deutschen Reich, Vol. 3, No. 8,
Aug., 1939.

WOOD

p. 30, American Museum of Natural History, N.Y.;
p. 31 (above left) courtesy Tennessee Valley Author-
ity, Knoxville, Tenn.; (above right) The Bettmann
Archive, N.Y.; (center right) Pope, Thomas: Treatise
on Bridge Architecture, N.Y., printed for the author
by A. Niven, 1811, pi. 9; (below) Museum of Modern
Art, N.Y.; p. 32, Chinese News Service Photos, from
Paul Guillumette, Inc., N.Y.; p. 33, American Museum
of Natural History, N.Y.; p. 34 (above and below)
courtesy Engineering News Record, N.Y.; (left)
Fletcher, Robert and Snow, J. P.; History of the
Development of Wooden Bridges, Paper #1864,

ASCE Transactions, N.Y., 1934; p. 35, Edmund H.
Royce, from Congdon, Herbert Wheaton: The Cov-
ered Bridge, N.Y., Knopf, 1946; p. 36 (above and
below) Pope, Thomas: Treatise on Bridge Architec-
ture, N.Y., printed for the author by A. Niven, 1811;
p. 37, courtesy C.L.V. Meeks, print in the William
Barclay Parsons Collection at Columbia University,
made from a drawing by G. A. Busby and pub-
lished in London by Taylor in 1823; p. 38, P. A.
Dearborn, N.Y.; p. 39, Fachklasse f Or Fotografie,
Gewerbeschule Zurich, courtesy Das V/erk.

METAL ARCH

p. 40 (above) Gauthey, Emiland Marie: Oeuvres
Traite de la Construction des Ponts, Navier, ed.,
Paris, Didot, 1813, Vol. II, pl.V-1; (center and be-
low) courtesy Henry-Russell Hitchcock, Jr., prints in
the Science Museum, London, made from drawings
by J. Raffield and published in London by Taylor in
1798; p. 41, courtesy Henry. Russell Hitchcock, Jr.,
from an aquatint of 1801 in the Science Museum,
London; p. 42 (above) Telford, Thomas: Life of
Thomas Telford, London, Payne and Foss, 1838,
(center) E. Jervoise, the National Buildings Record,
London; (below) courtesy American Bridge Com-
pany, Pittsburgh; p. 43, ND, courtesy L'Architecture
d Aujourd'hui, Paris; p. 44 (above) courtesy Hanover

and Hardesty, N.Y.; (center) Underwood and Under-
wood, courtesy American Institute of Steel Con-
struction, N.Y.; (below) Die Schweizerische Bau-
zeitung, 1913, courtesy Paul Zuberbuhler; p. 45,
Rodney McKay Morgan, N.Y.; p. 46 (below) courtesy
O. H. Ammann; p. 47, G. E. Kidder Smith, N.Y.;
p. 48, Hoyt, courtesy Port of New York Authority;
p. 49, courtesy Port of New York Authority; p. 50
(above) courtesy American Institute of Steel Con-
struction, N.Y.; (center and below) and p. 51, cour-
tesy State of New Hampshire Highway Department;
p. 52, courtesy American Institute of Steel Construc-
tion, N.Y.; p. 53 (above) courtesy Chicago Park Dis-
trict; (below) Hedrich-Blessing Studio, Chicago.

126

SUSPENSION CABLE

p. 55 (above) Navier, Claude: Rapport et Memoire sur
les Ponts Suspendus, Paris, Imprimerie royale, 1823,
pi. I; (below) courtesy Rudolf Mock; p. 56 (above)
Brunei, Isambard: life of I. K. Brunei, London, Long-
mans Green, 1870, frontis.; (center and below) The
Architectural Review, London, Sept., 1939, courtesy
The Central Library, Bristol; p. 57 (above) courtesy
Henry-Russell Hitchcock, Jr.; (below) R. Wills, the
National Buildings Record, London; p. 58, courtesy

New York Department of Public Works; p. 59 (above)
Keystone View Co., Inc., N.Y.; (below) courtesy Port
of New York Authority; p. 60 (above) courtesy Red-
wood Empire Association, San Francisco; (below)
Cine-photographie, Quebec, courtesy Quebec De-
partment of Public Works; p. 61, Rodney McKay
Morgan, N.Y.; p. 62, Kunst im Deutschen Reich,
Vol. 6, No. 12, Dec, 1942; p.63, Wide World Photos,
courtesy Engineering News Record, N.Y.

METAL BEAM

p. 64, courtesy Engineering News Record, N.Y., p. 65,
Schaechterle, Karl W. and Leonhardt, Fritz: Die
Gestaltung der Brucken, Berlin, Volk und Reich
Verlag, 1937; p. 66, Kunst im Deutschen Reich, Vol.
3, No. 8, Aug., 1939; p.67 (above and center) cour-
tesy Tennessee Valley Authority, Knoxville, Tenn.;
(below) courtesy Triborough Bridge Authority, N.Y.;
p. 68 (above and center) courtesy American Bridge
Company, Pittsburgh; (below) and p. 69, courtesy
Tennessee Valley Authority, Knoxville, Tenn.; p. 70
(above) courtesy American Institute of Steel Con-
struction, N.Y.; (below) courtesy Tennessee Valley
Authority, Knoxville, Tenn.; p. 71, Bureau of Recla-
mation, courtesy Engineering News Record, N.Y.;
p. 72 (above) Schaechterle, Karl W. and Leonhardt,
Fritz: Die Gestaltung der Brucken, Berlin, Volk und
Reich Verlag, 1937; (below) courtesy Port of New
York Authority; p.73, Foto Gross, St. Gallen O.,

Switzerland, courtesy Max Bill, Zurich; p. 74, British
Information Services, N.Y., courtesy Engineering
News Record, N.Y.; pp.75 and 76 (above and below)
courtesy Engineering News Record, N.Y.; p. 76 (cen-
ter) International Commercial Photo Co., N.Y., cour-
tesy American Institute of Steel Construction, N.Y.;
p. 77, courtesy Engineering News Record, N.Y.; p. 78
(above) courtesy I'Archifecfure d'Aujourd'hui, Paris;
(center) courtesy Engineering News Record, N.Y.;
(below) courtesy American Institute of Steel Con-
struction, N.Y.; pp.79, 80 (above) and 81, courtesy
Engineering News Record, NY.; p. 82, Ewing Gallo-
way, N.Y.; p. 83 (above) Mohringer, Karl: The Bridges
of the Rhine, Baden, Germany, Jon. Mohringer
Verlag, 1931, courtesy Engineering News Record,
N.Y.; (center) courtesy Engineering News Record,
N.Y.; (below) courtesy L' Architecture d'Aujourd'hui,
Paris.

REINFORCED CONCRETE

p. 84 (above) courtesy San Francisco Board of Park
Commissioners; (below) courtesy Engineering News
Record, N.Y.; p. 85 (above) courtesy National Park
Service, Washington, D.C.; (center) courtesy Con-
necticut Highway Department; (below) courtesy

Engineering News Record, N.Y.; p. 86, te Befon
Arme, Mar. 3, 1919, courtesy L Architecture d'Au-
jourd'hui; p. 87, Bureau Maillart, L. Meisser, Ing.,
courtesy Dr. Sigfried Giedion, Zurich.

REINFORCED CONCRETE ARCH

p. 88 (above) courtesy Cuyahoga County Engineer,
Ohio; (below) and p. 89, courtesy California Depart,
ment of Public Works; p. 90 (obove) Skansko Cement
Company, Malmo, Sweden, courtesy G. E. Kidder
Smith, N.Y.; (below) G. E. Kidder Smith, N.Y.; pp.91
and 92 (above) G. E. Kidder Smith, N.Y.; (below)
courtesy R. E. Enthoven, Librarian, Royal Institute of
Brifish Architects; p.93, G. E. Kidder Smith, N.Y.;
pp.94 and 95, courtesy Paul Zuberbuhler; p. 96,
H. Baranger, Paris, courtesy L'Archifecfure d'Aujour-
d'hui, Paris; pp.97 and 98, Skanska Cement Com-
pany, Malmo, Sweden, courtesy G. E. Kidder Smith,
N.Y.;p.99 (above) courtesy L' Architecture d'Aujour-

d'hui, Paris; (below) The Archifecfuro/ Forum, N.Y.;
pp.100 and 101, H. Baranger, Paris; p. 102, courtesy
Max Bill, Zurich; p. 103, O. Engler, Winterthur,
Switzerland, courtesy Dr. Sigfried Giedion, Zurich;
pp.104 and 105, Max Bill, Zurich; pp.106 and 107,
Mischol, Schiers, courtesy Dr. Sigfried Giedion,
Zurich; p. 108 (above) Max Bill, Zurich; (below) H.
Wolf-Bender's Erben, Zurich; p. 109, courtesy Dr.
Sigfried Giedion, Zurich, p. 110 (above) P. Bois-
sonas, Geneva; (center) H. Wolf-Bender's Erben,
Zurich; (below) Max Bill, Zurich; p.l 1 1, P. Boissonas,
Geneva.

REINFORCED CONCRETE: BEAM AND RIGID FRAME

p.l 12, Max Bill, Zurich; p.l 13, Ryner, courtesy
Prader & Cie, Zurich; pp.114 and 115, Sunami,
N.Y.; p.l 16 (above) courtesy Tennessee Valley Au-
thority, Knoxville, Tenn.; (upper center) courtesy
G. E. Kidder Smith, N.Y.; (lower center) courtesy
Engineering News Record, N.Y.; (below) courtesy
E.S. Elcock, p.l 17, courtesy Engineering News Rec-

ord, N.Y.; p.l 1 8 (above) F. Zurcher, Lausanne, cour-
tesy Paul Zuberbuhler; (below) "Topical'' Press
Agency, London, courtesy The Architectural Review,
London; p.l 19, Dell and Wainwright, courtesy The
Archifecfuro/ Review, London; pp.120 and 121,
Sunami, N.Y.

127

THIS BOOK WAS PRINTED IN 194?
FOR THE TRUSTEES OF THE MUSEUM OF MODERN ART
BY MODERN GRAVURE CORPORATION, NEW YORK
COVER AND TYPOGRAPHY BY EDWARD L. MILLS

MARSTON SCIENCE LIDRARY

Date Due

'^■- ■ ' '■■?■;

NOV 17 1997

Returned

-'

/

t

I

B

The architecture of bridges sci
624K19a

3 12bE 03213 MDDfl

MmiQH SCtEHCE UBftA**

\\

Mi

*

\ V

N*

■'*■■>
i ^

■\.

iW-WIWWfftttl' 'In 'I' ■■ ■ ■ ■ ■ **r~rrr~tiT*res